Search Results (23,668)

Background: Integrating human neural signals with computational vision systems offers a promising route toward more robust visual recognition, yet supporting mixed-granularity recognition, where both coarse- and fine-grained categories must be distinguished within a unified system, remains challenging due to the heterogeneous feature spaces of electroencephalography (EEG) and visual data. Methods: We propose “Align and Fuse,” a two-stage Transformer-based framework. Stage 1 constructs a shared semantic space using a hardness-aware multimodal supervised contrastive loss with Hard Negative Weighting to explicitly target confusable class pairs. Stage 2 employs a multimodal Transformer with co-attention to fuse the aligned features for classification. Results: On the 80-class EEG-ImageNet benchmark, our framework achieved 91.12% Top-1 accuracy under a temporally separated control protocol, improving over the corresponding vision-only (89.08%) and Standard Transformer (89.95%) baselines. Under the original stratified random split, it achieved 92.56% Top-1 accuracy; on the 40-class EEGCVPR dataset, accuracy reaches 95.82%. Cross-subject experiments yield 90.92% average Top-1 accuracy on four unseen subjects, and Grad-CAM analysis suggests that aligned EEG signals shift the model’s attention toward semantically relevant regions. Conclusions: Coupling hardness-aware alignment with decoupled multimodal fusion supports EEG-augmented recognition by leveraging complementary stimulus-related information under the evaluated protocols. Because EEG features are required at inference time, the framework is positioned as a human-in-the-loop EEG-augmented recognition system rather than a standalone vision model. Full article

Autonomous driving systems rely on a sophisticated pipeline of artificial intelligence models to perceive, predict, and plan in dynamic environments. This review presents a systematic analysis of the machine learning and deep learning models underpinning vehicle autonomy, spanning classical convolutional neural networks (CNNs) for object detection and semantic segmentation to recurrent and Transformer-based architectures for trajectory prediction and motion planning. It also provides a critical examination of the autonomous vehicle sensor stack, including cameras, LiDAR, radar, ultrasonics, and GNSS/IMU as data acquisition systems, highlighting modality-specific AI challenges such as monocular depth estimation, 3D point cloud processing, and radar Doppler interpretation. The evolution of perception and decision-making pipelines is reviewed, contrasting modular architectures with end-to-end learning paradigms that directly map raw sensor data to control commands, and discussing their trade-offs in interpretability, safety assurance, and robustness to rare edge cases. We further survey specialized hardware accelerators and heterogeneous automotive SoCs designed to meet stringent real-time and power constraints. Industrial strategies are compared, including multi-modal sensor fusion and vision-centric approaches based on large-scale imitation learning. Finally, we identify open challenges related to robustness under adverse conditions, domain shift, causal ambiguity, and the need for interpretable and certifiable AI in safety-critical autonomous driving systems. Full article

Background: Whether financial deepening and trade integration support or hinder environmental sustainability in hydrocarbon-dependent economies remains contested. Methods: This study examines the relationships among financial inclusion, energy intensity, trade openness, and CO₂ emissions per capita in Saudi Arabia for 1980–2020. The empirical strategy combines ARDL bounds testing, FMOLS, DOLS, CCR robustness, Toda–Yamamoto causality, and a battery of structural-break tests comprising Zivot–Andrews unit-root tests, Bai–Perron sup-F tests, and Chow tests. To address the mechanical correlation between carbon productivity and GDP, the per capita emissions specification (LNCP) is used as the primary outcome; carbon productivity (LNES) is reported for robustness. The small-sample sub-period results are stress-tested using ridge regression, residual-bootstrap confidence intervals, a GDP-augmented (scale-control) specification, and a break-date sensitivity analysis. Results: Cointegration is established. The Chow test identifies a significant break in the cointegrating relationship at 2001 (F = 7.36, p < 0.001 for LNCP), supported by the Zivot–Andrews endogenous-break dates for the financial-inclusion series (2000) and trade-openness series (2005), and by the Bai–Perron sup-F test (sup-F = 26.37 at 1990, exceeding the 1% Andrews critical value). Sub-sample re-estimation around 2001 shows that energy intensity, urbanisation, and trade openness are robust drivers of per capita emissions only after the break, while financial inclusion is statistically insignificant in both regimes once the GDP–carbon-productivity mechanical relationship is removed. Conclusions: The Saudi finance–environment relationship is structurally unstable, and policy assessments based on full-sample averages can be misleading. The evidence is best read as describing regime-dependent, conditional long-run associations rather than as identifying structural causal effects. By exposing the interactions, synergies, and trade-offs among financial deepening (SDG 8), energy efficiency (SDG 7), sustainable consumption and production (SDG 12), and climate action (SDG 13), the study shows how this descriptive quantitative evidence can inform—rather than directly identify—an instrument-level policy discussion. The findings are consistent with a Vision 2030 mix that prioritises energy efficiency and green-finance reform, with implications for SDG Targets 7.3, 8.10, 12.2, and 13.2 across oil-exporting economies. Full article

Traditional manual grading of fresh chili peppers suffers from inconsistent quality control and low efficiency. To meet the demand for accurate fruit shape recognition during the post-harvest stage, this study proposes an intelligent recognition method based on an improved DenseNet-121 network. This approach facilitates the application of machine vision in agricultural sorting equipment. DenseNet-121 serves as the backbone network. The Convolutional Block Attention Module (CBAM) is introduced to enhance feature focus on fruit shapes. A regularization strategy (Dropout = 0.3, weight decay = 1 × 10⁻⁴) and a cross-entropy loss function with label smoothing (LS = 0.1) are integrated to optimize decision boundaries. These configurations prevent the model from overfitting to hard training labels and yield a robust classification architecture. Experimental results demonstrate that the proposed model achieves a precision of 90.09%, a recall of 89.60%, an F1-score (the harmonic mean of precision and recall) of 89.53%, and an overall accuracy of 89.74%. The model contains 7.09 M parameters and requires a single-frame inference time of 7.35 ms. Comprehensive evaluations indicate that the proposed model achieves an optimal balance among environmental noise robustness, prediction accuracy, and computational efficiency. Consequently, by maintaining high fine-grained classification accuracy alongside a low memory footprint and rapid inference speed, the model demonstrates strong potential for real-time deployment on resource-constrained edge devices within actual agricultural optical sorting equipment. Full article

Monitoring the dynamic response of structures subjected to operational loads is a key component of structural health assessment, providing valuable information for safety evaluation and maintenance planning. In the last decade, video-based measurements have received growing attention for modal identification and damage detection applications, offering a promising alternative to traditional sensor-based approaches. Unlike conventional monitoring systems, which provide discrete measurements and often require extensive instrumentation, computer vision techniques enable dense, non-contact measurements while reducing installation costs and accessibility constraints. Moreover, Motion Magnification algorithms can be combined with computer vision-based identification techniques to amplify displacements within selected frequency ranges, facilitating the detection of low-amplitude structural vibrations. In this work, a semi-automated methodology for structural identification is presented and validated through two experimental applications involving vibrating systems monitored with commercial cameras. The proposed framework combines computer vision algorithms, Motion Magnification (MM), correlation analysis, and Principal Component Analysis (PCA), the latter being adopted as a noise-reduction and dimensionality-reduction tool to extract the most informative features from large sets of time-histories. In contrast to previous studies primarily focused on damage detection and frequency evolution tracking, the present work specifically investigates the influence of key user-defined parameters on the reliability of the identified frequencies and provides practical calibration guidelines for future applications. The methodology was validated against reference measurements obtained from an optical monitoring system and it successfully identified the natural frequencies of the analysed structures with errors ranging from 0.84% to 1.75%. Sensitivity analyses performed on the region of interest size and position, as well as on the correlation threshold, demonstrated the robustness of the proposed workflow. The results confirm that the proposed approach represents a reliable, low-cost, and minimally invasive alternative to conventional dynamic monitoring techniques, while providing practical recommendations for its implementation in real-world structural health monitoring applications. Full article

Video anomaly detection (VAD) aims to identify abnormal events in videos. Due to the lack of high-quality training data with detailed annotations, current VAD methods can only produce video-level predictions. To remedy this, several methods attempt to synthesize pseudo video anomalies. However, these methods suffer from low realism and coarse annotations, which limits their performance in real-world scenarios. In this paper, we propose a framework for unsupervised anomaly video generation from solely normal videos, leveraging VLMs to generate structured textual descriptions of anomalies conditioned on the perception of this video. Then, abnormal segments are synthesized using VLMs based on the synthetic textual descriptions. As our framework is highly controllable, video-level and region-level labels can be obtained to provide fine-grained annotations. On top of the synthetic data, we develop a fine-grained VAD network to simultaneously produce video-level, frame-level, and region-level predictions. Experiments show that our method achieves remarkable fine-grained VAD performance. Full article

Indoor companion, rehabilitation, logistics, laboratory transport, and service robot scenarios require mobile platforms that can follow a human operator safely and flexibly under lighting changes, occlusion, texture-poor corridors, and dynamic pedestrian environments. Vision-, LiDAR-, and UWB-based following systems can provide high perception capability, but their deployment cost, environmental dependence, and sensing complexity remain limiting factors for low-perception-dependence applications. This paper presents a passive following system for a Mecanum-wheel mobile platform based on gimbal posture perception and orthogonal odometry fusion. A rope-tensioned two-axis gimbal is mounted above a 300 mm × 300 mm × 150 mm omnidirectional chassis, and a six-axis inertial sensor installed at the top of the gimbal detects pitch and roll changes induced by user traction. A piecewise posture-to-velocity mapping model with a dead zone, saturation, low-pass filtering, and acceleration limiting converts the user’s traction intention into planar velocity commands in the vehicle coordinate frame. To reduce pose errors caused by Mecanum-wheel slip and discontinuous roller-ground contact, two orthogonal passive odometry wheels and inertial attitude estimation are fused to provide planar position feedback for closed-loop following. A prototype was implemented using an Infineon TRAVEO CYT4BB77 controller, TI DRV8701E motor drivers, six-axis IMUs, magnetic encoders, and an embedded display interface. Experiments evaluated attitude estimation accuracy, planar localization accuracy, passive following performance, gyroscope compensation, and open-loop/closed-loop following. The compensated attitude module achieved a static yaw drift of 0.45 deg/h and a dynamic attitude RMSE below 0.56 deg. Orthogonal odometry fusion produced an average positioning error of 3.8 mm over a 3000 mm linear displacement, reducing error by approximately 84.6% compared with pure Mecanum-wheel drive odometry. In a 5000 mm forward traction task, closed-loop following reduced the average distance error from 38.6 mm to 11.5 mm compared with open-loop attitude mapping. The results indicate that the proposed gimbal-orthogonal odometry architecture provides a compact, intuitive, and environment-robust solution for passive following on omnidirectional mobile platforms. Full article

Unmanned Aerial Vehicle (UAV)-based small object detection is a challenging computer vision task. It is constrained by two primary factors: UAV platforms have limited onboard computational resources, and high-altitude objects often have weak features that are easily overwhelmed by complex backgrounds. To address these challenges, we propose LUIM-YOLO. First, a Lightweight Multi-Scale Feature Enhancement (LMSFE) module integrates parallel multi-scale convolutions with attention to strengthen small and low-contrast object feature extraction. Second, an Adaptive Multi-Scale Bottleneck (AMSB) module enhances key semantic features of small objects and spatial correlation of medium-scale objects. Third, an Enhanced Cross-layer Compensation Feature Pyramid Network (ECC-FPN) constructs cross-level interaction pathways to improve small object position and scale perception. Experimental results on VisDrone2019 show that compared with YOLOv8n, LUIM-YOLO reduces parameters by 57% and improves mAP@50 by 12.9%. Additional full-validation-set PyTorch inference tests on NVIDIA Jetson Orin show that LUIM-YOLO achieves 88.19 ms/image in FP32, indicating a parameter-efficient accuracy-oriented design with edge deployment potential. Full article

This paper examines the association between external verification of sustainability reports and ESG performance of Saudi-listed firms from the years 2014–2021. With regard to the Saudi stock exchange (Tadawul) dataset consisting of 188 firm-year observations, it is concluded that external sustainability report verification and ESG performance are positively associated. This study constructs the premise that the enhancement of credibility and transparency of sustainability reports in turn fosters stakeholder confidence. This paper documents a positive association between voluntary assurance and ESG performance from an emerging market perspective, which broadens the scope of the ESG literature. This observation particularly justifies the need to endorse more assurance services in support of sustainable development and to strengthen the reporting frameworks and policies. The study results support the objectives of Vision 2030, specifically the pillars of promoting environmental sustainability, corporate transparency, and governance. The evidence aligning national goals to encourage transparency in corporate systems and sustainability in assurance services is the positive relationship between ESG and sustainability reporting assurance. Moreover, the results highlight Saudi Arabia’s dedication to the United Nations Sustainable Development Goals, specifically SDG 12 (Responsible Consumption and Production), and SDG 13 (Climate Action), as they underscore the role of assurance and disclosure practices in fostering sustainable business practices in Saudi Arabia. Full article

Background/Objectives: Manual Pap smear analysis for cervical cancer screening is limited by inter-observer variability, time constraints, and restricted expert availability. Although convolutional neural networks (CNNs) have automated cervical cell classification, they remain limited in modeling long-range spatial dependencies and often lack clinical interpretability. Methods: In this study, Vision Transformer (ViT) architectures were systematically optimized to enhance automated cervical cancer screening and improve interpretability. The Herlev dataset (917 images: 242 normal, 675 abnormal) was utilized to optimize ViT-Tiny, a lightweight ViT architecture designed for reduced computational complexity, through a comprehensive evaluation of augmentation strategies, class weighting, and hyperparameters. Results: The optimal configuration achieved a cross-validation accuracy of approximately 95% (94.89% for the best replicated configuration), in which random horizontal flipping and class weighting (0.7 × 1.3) were identified as most effective. Gradient-weighted Class Activation Mapping (Grad-CAM) analysis confirmed that model attention corresponded to clinically relevant morphological features, including nuclei regions, cell boundaries, and chromatin texture, which align with cytopathological criteria. Conclusions: These findings indicate that Vision Transformers can deliver accurate and interpretable decision support for cervical cancer screening by combining competitive classification performance with attention-based transparency relevant to medical AI. Further validation on larger, multi-center datasets remains necessary before clinical deployment. Full article

Image segmentation is a critical step in computer vision systems for phytosanitary diagnosis in cacao production. However, the reliability of classical segmentation methods remains insufficiently assessed under real field conditions, where images captured under non-standardized conditions are often affected by variable illumination, complex backgrounds, partial occlusions, and chromatic similarity between cacao pods and surrounding vegetation. This study compares global thresholding, K-means clustering, and GrabCut using 343 cocoa pod images captured in uncontrolled agricultural environments with non-standardized mobile devices; low-resolution images were retained to preserve external validity. Robustness was assessed on the full dataset using unsupervised structural metrics, including the segmented area ratio (AS), the largest component ratio (LCR), and the catastrophic failure rate (FC), while accuracy was validated on 50 manually annotated images using Intersection over Union (IoU). Wilcoxon signed-rank tests indicated statistically significant differences among methods. GrabCut achieved the best performance (IoU = 0.814), high structural coherence (LCR = 0.985), and a low catastrophic failure rate (FC = 1.7%). In contrast, K-means showed severe fragmentation and instability, whereas global thresholding was highly sensitive to illumination variability and complex backgrounds. These results indicate that GrabCut provides a robust training-free baseline for cocoa pod segmentation under uncontrolled field conditions, particularly for offline phytosanitary analysis where annotated datasets, supervised training, or GPU-based deployment are limited. Full article

Background/Objectives: Impacted third molars may adversely affect adjacent second molars, leading to pathological conditions such as external root resorption and dental caries. Accurate assessment of these interactions is important for treatment planning and clinical decision-making. Although cone-beam computed tomography (CBCT) provides detailed three-dimensional imaging, image interpretation remains challenging. Recent advances in artificial intelligence have enabled automated radiographic analysis using deep learning methods. Methods: This retrospective study included 162 CBCT scans obtained from patients aged 18–75 years. A total of 306 third molar–second molar units were evaluated. Based on radiographic findings, interactions were categorized as independent, contact, or resorption. Several deep learning architectures were developed and evaluated, including conventional convolutional neural networks (CNNs), attention-based CNNs, and Vision Transformer (ViT) models. Performance was assessed using standard classification metrics, and an ensemble approach was applied to improve predictive stability. Results: Attention-based and Transformer-based models generally outperformed conventional CNN architectures. These models achieved better discrimination among the defined classes and demonstrated superior overall performance. The ensemble model produced the most reliable results, achieving the highest macro-area under the curve (macro-AUC) values. Distinguishing contact cases from independent cases was the most challenging task, whereas resorption cases were identified more consistently across different models. Conclusions: Transformer-based deep learning models showed promising performance for CBCT-based assessment of third molar–second molar interactions. Ensemble learning further improved classification reliability and robustness. These findings suggest that artificial intelligence-assisted systems may support early detection of third molar-related pathological changes and contribute to more accurate radiological evaluation and clinical decision-making. Full article

Background/Objectives: Visual impairment is an underrecognized but functionally disabling complication of stroke that adversely affects rehabilitation potential, autonomy, and quality of life. Clinical, anatomical, and ophthalmologic determinants of post-stroke visual recovery remain incompletely defined, particularly in Eastern European tertiary stroke units where structured visual follow-up is not standardized. This study aimed to identify clinical, imaging, and ophthalmologic predictors of favorable visual recovery and to evaluate whether integrating these domains improves early prognostic stratification beyond standard neurological assessment. Methods: We conducted a retrospective single-center cohort study of 71 consecutive adult patients admitted with acute stroke and a documented visual complication between January 2022 and September 2025 at Pius Brinzeu Emergency County Hospital and Victor Babes University of Medicine and Pharmacy Timisoara. Favorable recovery was defined as ≥50% improvement in visual field index (VFI) at 6 months. Group comparisons used Student’s t-test, Mann–Whitney U test, chi-square test, and Fisher’s exact test. Multivariable logistic regression, Cox proportional hazards modeling, and unsupervised k-means clustering were performed. Results: Twenty-nine patients (40.8%) achieved favorable recovery, while 42 (59.2%) had persistent impairment. Responders were younger (62.8 ± 10.7 vs. 70.4 ± 10.8 years, p = 0.005) and had lower admission National Institutes of Health Stroke Scale (NIHSS) (6.4 ± 2.9 vs. 10.3 ± 4.4, p < 0.001), smaller lesion volumes (18.7 ± 11.4 vs. 33.2 ± 18.7 mL, p < 0.001), thicker peripapillary retinal nerve fiber layer (89.3 ± 7.6 vs. 78.2 ± 9.4 μm, p < 0.001), and earlier rehabilitation initiation (11.4 ± 5.3 vs. 21.7 ± 9.8 days, p < 0.001). NIHSS, time to rehabilitation, and optical coherence tomography (OCT) pRNFL thickness remained independent predictors. The full integrated model achieved an area under the receiver operating characteristic curve (AUC) of 0.87. Cluster analysis identified three distinct phenotypes with favorable recovery rates of 79.2%, 34.8%, and 8.3%. Conclusions: Combined clinical, neuroimaging, and ophthalmologic profiling—particularly OCT pRNFL—meaningfully refines early prediction of post-stroke visual recovery and supports phenotype-driven rehabilitation planning. Full article

The construction industry continues to face high levels of accidents despite the use of various safety training approaches, highlighting the need for more effective and responsive methods. This study examines the role of Generative Artificial Intelligence (GenAI) in potentially improving construction safety training by exploring the development of training practices and identifying the shortcomings of existing approaches. A systematic literature review (SLR) was conducted to analyse safety training methods and emerging GenAI applications, followed by validation interviews with industry experts in South Australia to ensure practical relevance. Emergent findings show that safety training has progressed through three main stages: instructor-led, digital and GenAI-enabled. However, instructor-led and digital approaches remain limited by non-interactive learning, limited flexibility to different learner needs, lack of real-time feedback and weak alignment with actual site conditions. In contrast, GenAI offers opportunities to support more interactive, personalised and context-aware training through technologies such as large language models (LLMs), adaptive learning systems, computer vision and scenario generation. Despite these benefits, significant challenges related to data quality, system reliability, ethical concerns and organisational readiness continue to affect implementation. Based on these findings, the study develops an integrated framework that links training evolution, key challenges and GenAI capabilities, providing practical guidance to improve safety training in construction. Full article

Automatic License Plate Recognition (ALPR) in unconstrained traffic environments requires simultaneously addressing two fundamental challenges: reliable localization of small and degraded license plates and accurate decoding of visually ambiguous character sequences. This paper presents a hierarchical multi-stage framework that combines deep-learning-based detection, geometric normalization, dual-channel recognition, and structured post-correction to improve recognition robustness under diverse real-world conditions. A systematic ablation study involving five configurations (A0–A4) demonstrates the effectiveness of the proposed architecture across three benchmark datasets. On the UC3M-LP dataset, exact-match accuracy increases from 45.2% to 88.3%, while achieving 91.6% partial accuracy and a zero detection-miss rate. The framework further attains 95% exact-match accuracy on controlled European license plate crops and 93% on a large-scale custom dataset. In addition, we identify systematic evaluation artifacts in partially annotated benchmarks, showing that truncated ground-truth labels can underestimate genuine character-level improvements. The proposed framework supports multiple license plate formats through a configurable structural template library, and preliminary experiments on a small Arabic-script subset suggest potential extensibility without full model retraining. To ensure full reproducibility, all source code and evaluation resources are publicly released. Full article