Search Results (1,759)

Underwater images often suffer from color distortion, low contrast, and structural blurring caused by wavelength-dependent absorption and scattering, which degrade both visual observation and downstream perception. Existing underwater image enhancement methods usually learn image-level restoration mappings, while the relationships among semantic regions, degradation patterns, and restoration responses are not fully exploited. In this paper, we propose a Semantic Guidance Network with Degradation Consistency for Underwater Image Enhancement (SGDC-UIE). Specifically, SGDC-UIE first extracts dense semantic responses from a frozen DINOv3 prior and converts them into foreground, boundary, and background region gates. These gates are then used to guide pseudo-physical degradation estimation, producing attenuation-like, transmission-like, illumination, structure, and background-light priors for region-aware restoration. These pseudo-physical priors are learned, bounded conditioning variables rather than calibrated estimates of underwater optical parameters. Based on these degradation conditions, a dual-branch restoration network corrects low-frequency color and illumination degradation while recovering high-frequency structural details through semantic-aware wavelet restoration. The color-restored and structure-restored outputs are further integrated by a degradation-consistent fusion gate, which adaptively balances visual fidelity and task-relevant structure preservation. In addition, grouped supervision with quality-anchor replay stabilizes task-aware fine-tuning and reduces visual-quality drift. Extensive experiments on paired and no-reference underwater enhancement benchmarks, semantic segmentation, and underwater object detection show that SGDC-UIE achieves competitive restoration quality and improves the usability of enhanced images for downstream perception. Full article

Correlated color temperature (CCT) can substantially influence image-based color measurements, though its effects vary with the optical and physical properties of the meat surface being evaluated. This study investigated the effects of CCT on the image-based color properties of skinless chicken breast and skin-on drumstick samples. It evaluated the suitability of 6500 K as a reference condition. Seventeen skinless breast and seventeen skin-on drumstick samples were repeatedly photographed under 13 CCT conditions ranging from 2500 to 8500 K at 500 K intervals, producing 442 observations. Camera settings, imaging geometry, illuminance, and sample position were maintained constant. Image-derived L*, a*, and b* values were obtained using the Fiji freeware program, and chroma (C*), whiteness index (WI), yellowness index (YI), browning index (BI), and total color difference (ΔE*ab) relative to 6500 K were calculated. Data were analyzed using linear mixed models, and the common variation in L*, a*, and b* was summarized by principal component analysis. Sample type, CCT, and their interaction significantly affected all evaluated parameters (p < 0.001). Increasing CCT generally increased L* and WI, while a*, b*, C*, YI, and BI decreased markedly, indicating a transition toward a lighter and less chromatic image-derived appearance. The largest deviations from the 6500 K reference occurred at 2500 K, with ΔE*ab values of 42.86 and 43.57 for breast and drumstick samples, respectively. Even at 6000 K, only 500 K below the reference condition, ΔE*ab exceeded 4.0 in both breast and drumstick samples, indicating visually perceptible color differences. Visually relevant differences also persisted above 6500 K. The first principal component explained 86.89% of the total variance and described the coordinated increase in lightness and reduction in redness and yellowness with increasing CCT. These findings demonstrate that CCT is a major source of systematic variability in image-based poultry color analysis workflow; therefore, 6500 K should not be treated as a universal reference unless the complete illumination, camera, calibration, and image-processing protocol is standardized and validated for the specific poultry surface being examined. Full article

Population aging has increased the need to improve visual comfort in age-friendly residential interiors. However, limited empirical evidence is available on the combined perceptual effects of correlated color temperature (CCT) and wood surface gloss in older adults. This study recruited 30 older adults and used a 2 × 3 within-subject design combining two CCT levels (2700 K and 4000 K) with three wood surface gloss levels (low, medium, and high). Eye-tracking, paired-image preference selection, and psychological impression ratings were used to assess visual attention, choice behavior, and subjective perception in a wood-based residential interior. The 4000 K condition produced greater sustained visual attention and a higher selection proportion than the 2700 K condition. However, the two CCT levels showed different subjective advantages: 2700 K received higher ratings for Warmth and Relaxation, whereas 4000 K performed better in Naturalness, Interest, Desire to use, and Overall comfort. Wood surface gloss affected visual attention and psychological impressions but showed no significant main effect on selection preference. Selected images also received consistently greater visual attention than unselected images. These findings indicate that CCT and wood surface gloss influence older adults’ visual responses through different perceptual dimensions. The results support function-specific coordination of lighting CCT and wood surface treatment in age-friendly residential interiors rather than a single uniform configuration. Full article

Contemporary lifestyles lead the elderly to spend up to 80% of their time indoors. Research underscores the unique visual requirements of different types of spaces to enhance accessible design for aging populations. This paper aims to evaluate the daylight conditions of the interior spaces of a senior living facility in Algeria. Interviews and on-site illuminance measurements were conducted to investigate perceptions and preferences regarding these spaces. A geometric model was then created in Rhino 8 and Grasshopper to simulate the interior daylight environment over an entire year. The study reveals significant daylighting deficiencies in the investigated spaces, including insufficient illuminance uniformity, localized overexposure near window zones with a potential glare risk, and some underlit activity areas. Additionally, annual daylight simulations revealed marked differences between the two shading scenarios. With the existing shading devices, the north-west-facing bedroom (R1NW) achieved UDI of 87.09%, whereas the south-east activity room (AR1SE) reached only 13.78%; removing the shading increased acceptable UDI in all investigated spaces. However, annual glare analysis showed that this improvement in daylight availability was accompanied by a reduction in glare autonomy. These results support orientation-specific daylighting strategies that include optimized or adaptable shading devices, supplementary task lighting for underlit functional areas, and increased interior finish reflectance to improve daylight distribution while reducing glare. Full article

This paper presents an explainable defense framework against perception-layer and Man-in-the-Middle (MitM) attacks in Internet of Things (IoT)-based environmental hazard warning systems. These systems rely on heterogeneous sensors (gas, light, sound, temperature, and humidity) whose integrity is crucial for reliable environmental alerts. Perception-layer attacks such as spoofing, jamming, and data injection can compromise sensor readings, while MitM attacks threaten communication reliability. The proposed approach integrates incremental Dynamic Time Warping (DTW) for time-series anomaly detection with a tree- based ensemble classifier (XGBoost), in addition to Shapley Additive Explanations (SHAP) for interpretability. A comparative evaluation framework jointly considers detection performance and explanation quality through metrics including pre-registering a Casual Ground Truth based on network protocol localized Precision@ K feature overlap metrics (Q), instead of relying on subjective human-expert or global rank correlations to quantitively evaluate the explanation transparency. Experimental simulations using an authentic EdgeIIoT-2022 dataset under 3-fold forward–chaining cross-validation demonstrated high detection accuracy and moderated explainability scores. The results prove the framework’s ability to detect and explain adversarial behaviors in sensor networks, strengthening trust, transparency, and resilience in safety-critical IoT infrastructures. Full article

Urban traffic congestion remains a major barrier to sustainable mobility, necessitating a transition from static signaling to Intelligent Traffic Light Systems (ITLS). This study presents a systematic review in computer vision, artificial intelligence, and adaptive control algorithms for urban intersection optimization. Following a methodological framework informed by PRISMA guidelines, the study examines state-of-the-art architectures, including deep learning-based perception models and reinforcement learning agents. The findings indicate that, although two-stage detectors provide benchmark accuracy for vehicle perception, single-stage models and Vision Transformers offer the high-speed processing required for real-time traffic management. In addition, deep reinforcement learning enables autonomous, lane-specific optimization that outperforms traditional actuated and fixed-time systems. The review also identifies a persistent research gap in the deployment of these computational frameworks in the resource-constrained and heterogeneous infrastructures of developing countries. For the urban context of Riobamba, Ecuador, a phased implementation strategy is proposed that balances computational demands with the city’s morphological and social characteristics. By bridging the gap between high-fidelity simulation and practical field deployment, this review provides a scalable framework for improving throughput, reducing emissions, and enhancing safety. Overall, these advances offer a promising pathway toward more resilient transportation systems in rapidly evolving Andean urban centers. Full article

Autonomous driving systems relieve drivers from continuous vehicle operation and constant monitoring, allowing them to engage in non-driving-related tasks (NDRTs). However, immersion in such tasks can impair drivers’ perception of both the takeover situation and the surrounding environment. To quantitatively assess drivers’ risk perception capability during takeover, a driving simulation platform was used to design autonomous takeover scenarios involving three types of NDRTs, three takeover request times (TOR), and two obstacle avoidance conditions. A total of forty participants were recruited to complete the driving experiment. Drivers’ eye movement data were collected, and visual metrics—including fixation, saccade, and pupil diameter—were extracted by defining areas of interest (AOIs). A subjective risk perception scale was developed and administered to measure drivers’ subjective evaluations. Together with takeover reaction time, the K-means clustering method was applied to classify drivers’ risk perception levels into three categories: high, medium, and low. The LightGBM algorithm was selected to construct a baseline classification model for assessing drivers’ risk perception levels. Subsequently, the Whale Optimization Algorithm (WOA) was employed to optimize the hyperparameters of LightGBM, resulting in the WOA-LightGBM model. This optimized model demonstrated improved recall, accuracy, precision, and F1-score, reaching 0.9210, 0.9253, 0.9261, and 0.9201, respectively. Furthermore, SHapley Additive exPlanations (SHAP) analysis was conducted to quantify the contribution of eye movement indicators to risk perception assessment. The results revealed that saccade duration in the NDRT areas significantly reduced drivers’ risk perception levels (SHAP value = −0.71), whereas increased saccade duration in the forward road area effectively restored drivers’ risk perception capability (SHAP value = 0.71). In addition, higher risk perception levels were found to enhance drivers’ takeover performance in terms of vehicle control. These findings provide valuable insights for the management of NDRTs and the optimization of autonomous vehicle takeover systems. Full article

[Background] Herpesvirus infections can induce diverse visual impairments with permanent sequelae, yet systematic data on their clinical spectrum and outcomes remain scarce. [Methods] We conducted a single-center retrospective cohort study at the Department of Neurology, Beijing Tongren Hospital, Capital Medical University. Thirteen consecutive patients (19 affected eyes) with herpesvirus-related visual impairment admitted between January 2016 and January 2025 were enrolled. Demographic data, clinical manifestations, etiological tests (polymerase chain reaction [PCR], metagenomic next-generation sequencing [mNGS], serology), neuroimaging, treatment regimens, and visual outcomes were analyzed. [Results] The cohort had a mean age of 50.4 years (range 31–66), with male predominance (84.6%, 11/13). Varicella zoster virus (VZV) was the leading pathogen (76.9%, 10/13), followed by herpes simplex virus type 1 (HSV-1), Epstein–Barr virus (EBV), and pseudorabies virus (PRV). Eight patients (61.5%) developed optic neuritis (ON) secondary to VZV infection, and five patients (38.5%) suffered from acute retinal necrosis (ARN), which was caused by VZV (n = 2), HSV-1 (n = 2), and PRV (n = 1). Bilateral involvement occurred in 46.2% (6/13) of patients. ARN was associated with the most severe visual loss. At the disease nadir, 46.2% of patients (6/13) presented with no light perception (NLP). Notably, five of these six NLP cases were diagnosed with ARN. Etiological confirmation was achieved in only 38.5% (5/13) of cases. mNGS of cerebrospinal and vitreous fluid, alongside aqueous humor PCR, are pivotal for diagnosing HSV-1/EBV mixed infections and rare PRV infection. All patients received antiviral therapy, 11 of whom (84.6%) were treated with intravenous antiviral agents. Glucocorticoids were administered as combination therapy to all patients. However, only one of eight VZV–ON eyes showed genuine visual improvement. In VZV–ARN, the initially involved eyes stayed NLP at final follow-up, while the fellow eyes recovered vision. Still, all non-VZV ARN patients had persistent bilateral NLP during follow-up. [Conclusions] Herpesvirus-associated visual impairment is dominated by VZV, manifests as ON or ARN, and carries a high risk of severe permanent vision loss—particularly in ARN. The emergence of zoonotic PRV underscores the need for heightened clinical vigilance. Diagnostic delays and insufficient interdisciplinary collaboration contribute substantially to poor outcomes. Full article

This study investigates the dynamics between media use, online activity, media literacy, and institutional trust among Indian citizens through a survey of adults distributed in six languages. Drawing on social capital theory, this study investigates how various media behaviors relate to perceptions of institutional actors such as governments, non-profits, and corporations as those actors increasingly seek digital avenues to reach constituencies. Findings underscore the significant relationships between media use, media literacy and positive views of democratic institutions, while online activities themselves did not appear to be significant. Notably, media literacy emerges as having a key positive tie to trust in political and state institutions as well as private, media, technology, and non-profit institutions. These results shed light on the nuanced relationship between digital engagement and perceptions of institutional credibility within the Indian context. Full article

Low-light image enhancement (LLIE) is a fundamental task in computer vision, required for restoring luminance, contrast, and structural fidelity in images captured under suboptimal lighting environments. This paper introduces an optimization-driven, scene-adaptive LLIE framework, designated as $O{M}_{\mathrm{I}\mathrm{D}\mathrm{C}\mathrm{P}}^{\mathrm{G}\mathrm{C}\mathrm{G}\mathrm{L}}$, which exploits the optical duality between low-light inversion and atmospheric scattering. The proposed methodology transforms low-light inputs into quasi-haze representations through an optical inversion process, followed by structural restoration using an Improved Dark Channel Prior ($M_{\mathrm{I}\mathrm{D}\mathrm{C}\mathrm{P}}$) baseline. To refine the restored output, a Gamma Correction with Global Limits (GCGL) module is integrated as a boundary constraint to mitigate localized over-exposure and preserve chromatic consistency. A core novelty of this framework lies in the deployment of metaheuristic optimization algorithms (MOAs)—specifically the Gray Wolf Optimizer (GWO), Harris Hawks Optimization (HHO), and Marine Predators Algorithm (MPA)—to autonomously resolve optimal, image-specific parameter configurations. This search paradigm is guided by perception-driven fitness functions, namely the Patch-based Contrast Quality Index (PCQI) or the Blind/Referenceless Image Spatial Quality Evaluator (BRISQUE). Quantitative and qualitative evaluations across a comprehensive pool of 1092 benchmark images demonstrate that the proposed framework exhibits robust statistical resilience and cross-dataset generalization compared to four state-of-the-art deep learning methods. While data-driven deep learning architectures retain localized superiority under the extreme degradation boundaries of the DARK FACE dataset, the proposed physics-inspired optimization framework achieves the leading overall cross-dataset aggregate ranking ($\overline{R}=2.467$) across diverse evaluation environments due to its per-image dynamic solution space mapping. Full article

Ground segmentation is an essential preprocessing step for LiDAR-based perception in autonomous driving, as it directly affects subsequent object clustering and obstacle detection. This paper proposes a fast grid-based ground segmentation method for 3D Light Detection and Ranging (LiDAR) point clouds using dual-seed expansion. The input point cloud is projected onto a two-dimensional Cartesian grid, where cell-wise height statistics, including minimum height, maximum height, mean height, and height variance, are computed. Initial ground candidates are selected by combining a global height-distribution-based seed and a LiDAR-mounting-height-based seed to reduce missed ground candidates in sloped or sparse regions. Ground cells are then expanded to nearby cells based on the local height difference. To reduce misclassification in cells containing both ground and non-ground points, a local height-band criterion is applied for point-level refinement. The proposed method was implemented in Robot Operating System 1 (ROS 1) and evaluated on the SemanticKITTI dataset. Experimental results showed an average accuracy of 96.34%, an F1-score of 97.11%, and a ground Intersection over Union (IoU) of 94.40%. The average processing time over SemanticKITTI sequences 00–10 was 12.15 ms per frame. These results demonstrate that the proposed method is suitable as a fast preprocessing module for LiDAR-based autonomous driving perception. Full article

Rapid preliminary reconnaissance of Critical Transportation Hubs (CTHs) by UAV swarms is vital during post-disaster rescue operations. However, limited camera fields of view, large heading variances, and GNSS multipath errors near massive steel-concrete structures complicate multi-view cooperative perception. This paper introduces a discrete, vision-based cooperative perception framework utilizing a decentralized anchor-wingman architecture. The pipeline integrates a Prob-IoU-optimized YOLO26m-OBB detector to extract oriented infrastructure footprints. To handle severe rotational discrepancies without IMU priors, a global scene registration cascade—combining SuperPoint and an Optimal Transport-driven LightGlue—is employed to establish robust geometric correspondences. Furthermore, a Projected Polygon Intersection over Union (Proj-IoU) mechanism, coupled with an RMSE-weighted spatial fusion strategy, dynamically associates and deduplicates overlapping targets across distributed views. Experimental results indicate that the framework achieves a low pixel-level RMSE of 2.12 pixels on the source domain and maintains a highly stable 2.36 pixels during zero-shot cross-domain testing (SUES-200 dataset), successfully resolving extreme heading variances up to 270°. The Proj-IoU mechanism resolves multi-source redundancies—collapsing overlapping projections by over 50%—bounding the localization error to approximately 1.06 m. Operating at 6.7 FPS on edge hardware via low-bandwidth tensor transmission, this system provides a rigorous geometric foundation for autonomous swarms, enabling downstream collision-free trajectory planning and Multi-Target Task Allocation (MTTA) in GNSS-denied environments. Full article

Traffic accidents involving pedestrians during the hours of darkness pose a serious threat to road safety due to reduced visibility and drivers’ delayed perception of the traffic situation. Accurate estimation of the detection distance for pedestrians is essential in the reconstruction of traffic accidents. This study analyzes the relationship between driver visibility, environmental conditions, and the ability to detect pedestrians in a timely manner during nighttime driving. The study examines the main factors influencing the driver’s “perception–reaction” process, including the illumination provided by the vehicle’s headlights, the illumination of the road environment, the contrast and reflective properties of the pedestrian’s clothing, as well as the driver’s level of attention. Using a graph-analytical method, the detection distances for pedestrians under nighttime conditions are estimated. A real-life accident scenario was reconstructed to determine whether the driver had sufficient time and distance to perceive the danger and take action to avoid a collision. The results show that pedestrian visibility depends on lighting conditions, which directly affect the driver’s reaction time. These findings contribute to the refinement of the methodological approach to reconstructing traffic accidents and can assist experts in conducting automotive technical examinations. Full article

Indoor fire smoke degrades visible-light cameras and near-infrared Lidar through wavelength-dependent absorption and scattering, threatening robotic navigation safety. Existing path planners either ignore sensor degradation or rely on empirical penalties lacking a physical basis. To address these issues, this paper proposes Quality Cost A* (QC-A*), which maps Fire Dynamics Simulator (FDS) visibility fields to sensor perception quality via the Koschmieder and Beer–Lambert physical laws, embedding a cost function that drives paths away from high-attenuation regions. A multi-sensor fusion layer provides fault tolerance under sensor-specific failure conditions. The method is validated through FDS-based simulations across four smoke scenarios in a $20\mathrm{m}\times 6\mathrm{m}$ corridor with 21 obstacles, using 50 start–goal pairs per scenario. Perception quality derives from Beer–Lambert optical transmittance, while the hazard-zone proportion quantifies path segments with visibility below 5 m. Across the Symmetric and Asymmetric scenarios, QC-A* reduces the low-visibility hazard-zone proportion from 40.7% to 19.6% and improves worst-case perception quality from 0.067 to 0.177, with a 15.3% path length increase, while remaining close to traditional A* in light-smoke conditions. Under constructed sensor failure tests, QC-A* maintains a 96–100% planning success rate versus 48% for Camera-Only and 70% for Lidar-Only. QC-A* shifts sensor degradation modeling from empirical penalty to physical mechanism, achieving a favorable safety–efficiency balance prioritizing perceptual safety, and provides an interpretable, generalizable framework for robotic fire-environment path planning. Full article

As artificial intelligence (AI) rapidly permeates entrepreneurial ecosystems, understanding how technological literacy relates to entrepreneurial team well-being has become an urgent priority. Drawing on conservation of resources (COR) theory, this study examines the relationship between entrepreneurial teams’ AI literacy and team well-being (distinguishing between hedonic and eudaimonic well-being). Furthermore, it investigates the mediating roles of entrepreneurial team efficacy and collective AI anxiety. Data were collected from a survey of 271 entrepreneurial teams across four major economic zones in China. This study relies on team leaders as primary informants to report team-level perceptions. The results show that AI literacy is positively related to team hedonic well-being, but exhibits no significant direct relationship with team eudaimonic well-being. However, AI literacy is indirectly associated with entrepreneurial team well-being through two parallel pathways: entrepreneurial team efficacy and collective AI anxiety. The findings shed light on how technological literacy is linked to team mental health via the dual mechanisms of cognitive resource gain and emotional loss prevention. This study broadens our understanding of well-being at the entrepreneurial team level and offers insights into cultivating literacy, enhancing efficacy, and managing emotions to improve entrepreneurial team well-being. Full article