Search Results (224)

The application of three-dimensional ground-penetrating radar (3D GPR) for intelligent pavement defect analysis is often constrained by the limited availability of labeled samples. To address this challenge, this study employed Ground Penetrating Radar Maxwell (GprMax) to simulate typical pavement defects, including cracks, loose materials, and interlayer debonding. A Cycle-Consistent Generative Adversarial Network (Cycle-GAN) was then introduced to perform style transfer on the simulated images, thereby reducing the domain gap between simulated and real radar images. Furthermore, four You Only Look Once (YOLO) models—YOLO version 5, YOLOX, YOLO version 7, and YOLO version 8—were systematically compared using real datasets to identify the best-performing model, which was subsequently used to evaluate the effect of different proportions of synthetic data on detection performance. The results demonstrated that the moderate inclusion of synthetic data improved the recognition accuracy of loose defects (from 76.7% to 78.9%), whereas its impact on crack and debonding detection was negative. Moreover, excessive reliance on synthetic data led to overfitting, thereby reducing the model’s generalization capability. Among the four models, YOLOv7 achieved the best overall performance, with a mean Average Precision (mAP) of 83.4% and a crack detection rate of 88.2%. This study thus provides a feasible technical pathway and model selection reference for automated GPR-based pavement defect identification, offering practical value for efficient and accurate road maintenance inspections. Full article

Public transportation is the core of easing urban traffic congestion, reducing pollution and advancing smart city transportation intellectualization. Its refined operation relies heavily on accurate, real-time passenger origin–destination (OD) data. However, traditional manual surveys are costly with low sampling rates, while smart card big data lacks alighting information and has deviations, failing to reflect real travel behaviors and becoming a bottleneck for intelligent public transportation development. To address this, this paper proposes a bus passenger boarding/alighting detection and recognition study based on video images and the YOLO algorithm. Aiming at traditional YOLO’s shortcomings in on-vehicle scenarios (insufficient feature extraction, inefficient feature fusion, slow convergence), the baseline YOLOv8n is improved for bus scenarios’ high-density, high-occlusion and variable-target scales: (1) DAC2f structure (deformable attention + C2f) captures occluded passengers’ core features and suppresses background interference; (2) SWD-PAN enables bidirectional cross-scale feature interaction to adapt to scale differences; and (3) WIoUv3 balances sample weights for small targets and non-standard posture passengers. Experiments show that precision, recall and mAP increase by 3.68%, 5.12% and 6.26%, respectively, meeting real-time requirements. The improved YOLOv8 is deeply integrated with DeepSORT to enhance tracking stability. Tests show that MOTA reaches 31.24% (2.6% higher than YOLOv8n, 16.4% higher than YOLO-X) and MOTP reaches 88.06%, solving trajectory breakage and ID switching. This addresses traditional OD data collection pain points, providing technical support for intelligent public transportation refined management and smart city transportation optimization. Full article

Synthetic Aperture Radar (SAR) is a powerful observation system capable of delivering high-resolution imagery under variable sea conditions to support target detection and tracking, such as for ships. However, conventional optical target detection models are typically engineered for complex optical imagery, leading to limitations in accuracy and high computational resource consumption when directly applied to SAR imagery. To address this, this paper proposes a lightweight shape-aware and direction-weighted algorithm for SAR ship detection, SADW-Det. First, a lightweight streamlined backbone network, LSFP-NET, is redesigned based on the YOLOX architecture. This achieves reduced parameter counts and computational burden by incorporating depthwise separable convolutions and factorized convolutions. Concurrently, a parallel fusion module is designed, leveraging multiple small-kernel depthwise separable convolutions to extract features in parallel. This approach maintains accuracy while achieving lightweight processing. Furthermore, addressing the differences between SAR imagery and other imaging modalities, a direction-weighted attention was devised. This enhances model performance with minimal computational overhead by incorporating positional information while preserving channel data. Experimental results demonstrate superior detection accuracy compared to existing methods on three representative SAR datasets, SSDD, HRSID and DSSDD, while achieving reduced parameter counts and computational complexity, indicating strong application potential and laying the foundation for cross-modal applications. Full article

An advanced artificial intelligence (AI)-powered mobile automated impurity removal system was developed and integrated into potato harvesting machinery for decentralized agricultural environments in Japan. As opposed existing stationary AI systems in centralized processing facilities, this mobile prototype enables on-field impurity removal in real time through a systematic dual-evaluation methodology. The system integrates the YOLOX-small architecture with precision pneumatic actuators and achieves 40–50 FPS processing under dynamic field conditions. Algorithm validation across 10 morphologically diverse potato varieties (Danshaku, Harrow Moon, Hokkaikogane, Kitaakari, Kitahime, May Queen, Sayaka, Snowden, Snow March, and Toyoshiro) using count-based analysis showed exceptional recognition, with potato misclassification rates of 0.08 ± 0.03% (range: 0.01–0.32%) and impurity detection rates of 89.99 ± 1.25% (range: 80.00–93.30%). Cross-farm validation across seven commercial farms in Hokkaido confirmed robust algorithm consistency (PMR: 0.08 ± 0.03%, IDR: 90.56 ± 0.82%) without farm-specific calibration, establishing variety-independent and environment-independent operation. Field validation using weight-based analysis during actual harvesting at 1–4 km/h confirmed successful AI-to-field translation, with 0.22–0.42% potato misclassification and adaptive impurity removal of 71.43–85.29%. The system adapted intelligently, employing conservative sorting under high-impurity loads (71.43% removal, 0.33% misclassification) to prioritize potato preservation while maximizing efficiency under standard conditions (85.29% removal, 0.30% misclassification). The dual-evaluation framework successfully bridged the gap between AI accuracy in laboratory settings and effectiveness in agricultural operations. The proposed AI algorithm surpassed project targets for all tested conditions (>60% impurity removal, <1% potato misclassification). This successful integration demonstrates technical feasibility and commercial viability for widespread agricultural automation, with a validated 50% reduction in labor (four workers to two workers). This implementation provides a comprehensive validation methodology for next-generation autonomous harvesting systems. Full article

Synthetic Aperture Radar (SAR) ship detection remains challenging due to background clutter, target sparsity, and fragmented or partially occluded ships, particularly at small scales. To address these issues, we propose the Deformable Recurrent Criss-Cross Attention Network (${\mathrm{DRC}}^2$-Net), a lightweight and efficient detection framework built upon the YOLOX-Tiny architecture. The model incorporates two SAR-specific modules: a Recurrent Criss-Cross Attention (RCCA) module to enhance contextual awareness and reduce false positives and a Deformable Convolutional Networks v2 (DCNv2) module to capture geometric deformations and scale variations adaptively. These modules expand the Effective Receptive Field (ERF) and improve feature adaptability under complex conditions. DRC²-Net is trained on the SSDD and iVision-MRSSD datasets, encompassing highly diverse SAR imagery including inshore and offshore scenes, variable sea states, and complex coastal backgrounds. The model maintains a compact architecture with 5.05 M parameters, ensuring strong generalization and real-time applicability. On the SSDD dataset, it outperforms the YOLOX-Tiny baseline with AP@50 of 93.04% (+0.9%), ${\mathrm{AP}}_s$ of 91.15% (+1.31%), ${\mathrm{AP}}_m$ of 88.30% (+1.22%), and ${\mathrm{AP}}_l$ of 89.47% (+13.32%). On the more challenging iVision-MRSSD dataset, it further demonstrates improved scale-aware detection, achieving higher AP across small, medium, and large targets. These results confirm the effectiveness and robustness of ${\mathrm{DRC}}^2$-Net for multi-scale ship detection in complex SAR environments, consistently surpassing state-of-the-art detectors. Full article

The detection of metallic surface defects is an essential task to control the quality of industrial products. During the production of metal materials, several defect types may appear on the surface, accompanied by a large amount of background texture information, leading to false or missing detections during small-defect detection. Computer vision is a crucial method for the automatic detection of defects. Yet, this remains a challenging problem, requiring the continuous development of new approaches and algorithms. Furthermore, many industries require fast and real-time detection. In this paper, a lightweight deep learning model is presented for implementation on embedded devices to perform in real time. The YOLOX-Tiny model is used for detecting and classifying metallic surface defect types. The YOLOX-Tiny has 5.06M parameters and only 6.45 GFLOPs, yet performs well, even with a smaller model size than its counterparts. Extensive experiments on the dataset demonstrate that the proposed model is robust and can meet the accuracy requirements for metallic defect detection. Full article

Marine concrete structures are continuously exposed to harsh marine environments—salt, waves, and biological fouling—that accelerate corrosion and cracking, increasing maintenance costs. Traditional Non-Destructive Testing (NDT) techniques often fail to detect early damage due to signal attenuation and noise in underwater conditions. This study critically reviews recent advances in Artificial Intelligence-integrated NDT (AI-NDT) technologies for marine concrete, focusing on their quantitative performance improvements and practical applicability. To be specific, a systematic comparison of vision-based and signal-based AI-NDT techniques was carried out across reported field cases. It was confirmed that the integration of AI improved detection accuracy by 17–25%, on average, compared with traditional methods. Vision-based AI models such as YOLOX-DG, Cycle GAN, and MSDA increased mean mAP 0.5 by 4%, while signal-based methods using CNN, LSTM, and Random Forest enhanced prediction accuracy by 15–20% in GPR, AE, and ultrasonic data. These results confirm that AI effectively compensates for environmental distortions, corrects noise, and standardizes data interpretation across variable marine conditions. Lastly, the study highlights that AI-enabled NDT not only automates data interpretation but also establishes the foundation for predictive and preventive maintenance frameworks. By linking data acquisition, digital twin-based prediction, and lifecycle monitoring, AI-NDT can transform current reactive maintenance strategies into sustainable, intelligence-driven management for marine infrastructure. Full article

The accurate assessment of Chinese bayberry (Myrica rubra) maturity is critical for intelligent harvesting. This study proposes a novel cascaded framework combining instance segmentation and multi-feature regression for accurate maturity detection. First, a lightweight SOLOv2-Light network is employed to segment each fruit individually, which significantly reduces computational costs with only a marginal drop in accuracy. Then, a multi-feature extraction network is developed to fuse deep semantic, color (LAB space), and multi-scale texture features, enhanced by a channel attention mechanism for adaptive weighting. The maturity ground truth is defined using the a*/b* ratio measured by a colorimeter, which correlates strongly with anthocyanin accumulation and visual ripeness. Experimental results demonstrated that the proposed method achieves a mask mAP of 0.788 on the instance segmentation task, outperforming Mask R-CNN and YOLACT. For maturity prediction, a mean absolute error of 3.946% is attained, which is a significant improvement over the baseline. When the data are discretized into three maturity categories, the overall accuracy reaches 95.51%, surpassing YOLOX-s and Faster R-CNN by a considerable margin while reducing processing time by approximately 46%. The modular design facilitates easy adaptation to new varieties. This research provides a robust and efficient solution for in-field bayberry maturity detection, offering substantial value for the development of automated harvesting systems. Full article

Purpose: To develop and validate a two-stage system for automated quality assessment of shoulder true-AP radiographs by combining joint localization with quality classification. Materials and Methods: From the MURA “SHOULDER” subset, 2956 anteroposterior images were identified; 59 images with negative–positive inversion, excessive metallic implants, extreme exposure, or presumed fluoroscopy were excluded, yielding a class-balanced set of 2800 images (1400 OK/1400 NG). A YOLOX-based detector localized the glenohumeral joint, and classifiers operated on both whole images and detector-centered crops. To enhance interpretability, we integrated Grad-CAM into both whole-image and local classifiers and assessed attention patterns against radiographic criteria. Results: The detector achieved AP@0.5 = 1.00 and a mean Dice similarity coefficient of 0.967. The classifier attained AUC = 0.977 (F1 = 0.943) on a held-out test set. Heat map analyses indicated anatomically focused attention consistent with expert-defined regions, and coverage metrics favored local over whole-image models. Conclusions: The two-stage, XAI-integrated approach provides accurate and interpretable assessment of shoulder true-AP image quality, aligning model attention with radiographic criteria. Full article

Conflicts in freeway interchange diverging areas remain poorly understood, particularly their characteristics and severity determinants. To address this gap, we extracted over 20,000 vehicle trajectories from UAV footage at 16 interchange divergence zone across five multi-lane expressways using a YOLOX–DeepSORT method. From these trajectories, we identified longitudinal and lateral conflicts and classified their severity into minor, moderate, and severe levels using a two-dimensional extended time-to-collision metric. Subsequently, we incorporated 19 macroscopic traffic-flow and microscopic driver-behavior variables into four conflict-severity models–multivariate logistic regression, random forest, CatBoost, and XGBoost—and conducted to identify the key determinants of conflict severity based on the optimal models. The results indicate that lateral conflicts last longer and pose higher collision risks than longitudinal ones. Furthermore, moderate conflicts are most prevalent, whereas severe conflicts are concentrated within 300 m upstream of exit ramps. Specifically, for longitudinal conflicts, the most influential factors include speed difference, target-vehicle speed, truck involvement, traffic density, and exit behavior. In contrast, for lateral conflicts, the most critical factors include lane-change frequency, speed difference, target-vehicle speed, distance to the exit ramp, and truck proportion. Overall, these findings support the development of hazardous-driving warning systems and proactive safety management strategies in interchange diverging areas. Full article

Iris recognition is currently considered the most promising biometric method and has been applied in many fields. Current commercial and research systems typically use software solutions running on a dedicated computer, whose power consumption, size and price are considerably high. This paper presents a hardware-based embedded solution for real-time iris segmentation. From an algorithmic point of view, the system consists of two steps. The first employs a YOLOX trained to detect two classes: eyes and iris/pupil. Both classes intersect in the last of the classes and this is used to emphasise the detection of the iris/pupil class. The second stage uses a lightweight U-Net network to segment the iris, which is applied only on the locations provided by the first stage. Designed to work in an Iris At A Distance (IAAD) scenario, the system includes quality parameters to discard low-contrast or low-sharpness detections. The whole system has been integrated on one MultiProcessor System-on-Chip (MPSoC) using AMD’s Deep learning Processing Unit (DPU). This approach is capable of processing the more than 45 frames per second provided by a 16 Mpx CMOS digital image sensor. Experiments to determine the accuracy of the proposed system in terms of iris segmentation are performed on several publicly available databases with satisfactory results. Full article

With growing global attention on animal welfare and food safety, humane and efficient slaughtering methods in the poultry industry are in increasing demand. Traditional manual inspection methods for stunning broilers need significant expertise. Additionally, most studies on electrical stunning focus on white broilers, whose optimal stunning conditions are not suitable for red-feathered Taiwan chickens. This study aimed to implement a smart, safe, and humane slaughtering system designed to enhance animal welfare and integrate an IoT-enabled vision system into slaughter operations for red-feathered Taiwan chickens. The system enables real-time monitoring and smart management of the poultry stunning process using image technologies for dynamic object tracking recognition. Focusing on red-feathered Taiwan chickens, the system applies dynamic tracking objects with chicken morphology feature extraction based on the YOLO-v4 model to accurately identify stunned and unstunned chickens, ensuring compliance with animal welfare principles and improving the overall efficiency and hygiene of poultry processing. In this study, the dynamic tracking object recognition system comprises object morphology feature detection and motion prediction for red-feathered Taiwan chickens during the slaughtering process. Images are firsthand data from the slaughterhouse. To enhance model performance, image amplification techniques are integrated into the model training process. In parallel, the system architecture integrates IoT-enabled modules to support real-time monitoring, sensor-based classification, and cloud-compatible decisions based on collections of visual data. Prior to image amplification, the YOLO-v4 model achieved an average precision (AP) of 83% for identifying unstunned chickens and 96% for identifying stunned chickens. After image amplification, AP improved significantly to 89% and 99%, respectively. The model achieved and deployed a mean average precision (mAP) of 94% at an IoU threshold of 0.75 and processed images at 39 frames per second, demonstrating its suitability for IoT-enabled real-time dynamic tracking object recognition in a real slaughterhouse environment. Furthermore, the YOLO-v4 model for poultry slaughtering recognition in transient stability, as measured by training loss and validation loss, outperforms the YOLO-X model in this study. Overall, this smart slaughtering system represents a practical and scalable application of AI in the poultry industry. Full article

The sheaths of bamboo shoots, characterized by distinct colors and spotting patterns, are key phenotypic markers influencing species classification, market value, and genetic studies. This study introduces YOLOv8-BS, a deep learning model optimized for detecting these traits in Chimonobambusa utilis using a dataset from Jinfo Mountain, China. Enhanced by data augmentation techniques, including translation, flipping, and contrast adjustment, YOLOv8-BS outperformed benchmark models (YOLOv7, YOLOv5, YOLOX, and Faster R-CNN) in color and spot detection. For color detection, it achieved a precision of 85.9%, a recall of 83.4%, an F1-score of 84.6%, and an average precision (AP) of 86.8%. For spot detection, it recorded a precision of 90.1%, a recall of 92.5%, an F1-score of 91.1%, and an AP of 96.1%. These results demonstrate superior accuracy and robustness, enabling precise phenotypic analysis for bamboo germplasm evaluation and genetic diversity studies. YOLOv8-BS supports precision agriculture by providing a scalable tool for sustainable bamboo-based industries. Future improvements could enhance model adaptability for fine-grained varietal differences and real-time applications. Full article

In this paper, we presents a case study involving the implementation experience and a methodological framework through a comprehensive comparative analysis of the YOLOX and YOLOv12 object detection models for agricultural automation systems deployed in the Jetson AGX Orin edge computing platform. We examined the architectural differences between the models and their impact on detection capabilities in data-imbalanced potato-harvesting environments. Both models were trained on identical datasets with images capturing potatoes, soil clods, and stones, and their performances were evaluated through 30 independent trials under controlled conditions. Statistical analysis confirmed that YOLOX achieved a significantly higher throughput (107 vs. 45 FPS, p < 0.01) and superior energy efficiency (0.58 vs. 0.75 J/frame) than YOLOv12, meeting real-time processing requirements for agricultural automation. Although both models achieved an equivalent overall detection accuracy (F1-score, 0.97), YOLOv12 demonstrated specialized capabilities for challenging classes, achieving 42% higher recall for underrepresented soil clod objects (0.725 vs. 0.512, p < 0.01) and superior precision for small objects (0–3000 pixels). Architectural analysis identified a YOLOv12 residual efficient layer aggregation network backbone and area attention mechanism as key enablers of balanced precision–recall characteristics, which were particularly valuable for addressing agricultural data imbalance. However, NVIDIA Nsight profiling revealed implementation inefficiencies in the YOLOv12 multiprocess architecture, which prevented the theoretical advantages from being fully realized in edge computing environments. These findings provide empirically grounded guidelines for model selection in agricultural automation systems, highlighting the critical interplay between architectural design, implementation efficiency, and application-specific requirements. Full article

Low-altitude drones, which are unimpeded by traffic congestion or urban terrain, have become a critical asset in emergency rescue missions. To address the current lack of emergency rescue data, UAV aerial videos were collected to create an experimental dataset for action classification and localization annotation. A total of 5082 keyframes were labeled with 1–5 targets each, and 14,412 instances of data were prepared (including flight altitude and camera angles) for action classification and position annotation. To mitigate the challenges posed by high-resolution drone footage with excessive redundant information, we propose the SlowFast-Traffic (SF-T) framework, a spatio-temporal sequence-based algorithm for recognizing traffic accident rescue actions. For more efficient extraction of target–background correlation features, we introduce the Actor-Centric Relation Network (ACRN) module, which employs temporal max pooling to enhance the time-dimensional features of static backgrounds, significantly reducing redundancy-induced interference. Additionally, smaller ROI feature map outputs are adopted to boost computational speed. To tackle class imbalance in incident samples, we integrate a Class-Balanced Focal Loss (CB-Focal Loss) function, effectively resolving rare-action recognition in specific rescue scenarios. We replace the original Faster R-CNN with YOLOX-s to improve the target detection rate. On our proposed dataset, the SF-T model achieves a mean average precision (mAP) of 83.9%, which is 8.5% higher than that of the standard SlowFast architecture while maintaining a processing speed of 34.9 tasks/s. Both accuracy-related metrics and computational efficiency are substantially improved. The proposed method demonstrates strong robustness and real-time analysis capabilities for modern traffic rescue action recognition. Full article