Search Results (12,263)

In night-time construction scenarios of under-construction nuclear power plants, some yellow lights and open flames exhibit highly similar visual characteristics, resulting in frequent false alarms of fire sources. Such false alarm information tends to drown out real fire alarm signals, which not only severely disrupts construction operations but also endangers fire safety. To address this problem, this paper proposes an intelligent fire risk identification method based on an enhanced YOLOv8n (named YOLO-Fire). Specifically, shallow convolutional layers embedded with a coordinate attention mechanism are integrated into the Backbone of YOLOv8n; the Neck is optimised to improve the efficiency of multi-scale feature fusion; and the Head is enhanced to strengthen the localization and classification branches. Additionally, a composite loss function combining classification loss, regression loss, and similarity loss is designed, coupled with night-scene-specific data augmentation techniques and a two-stage progressive training strategy. Experimental results show that YOLO-Fire reduces the false alarm rate by 14.3%, increases the mean average precision (mAP@0.5) for open flames by 11.3% to 75.2%, and maintains an inference speed of over 85 frames per second (FPS). This study achieves an optimal balance between false alarm control, small object detection accuracy, and real-time processing efficiency, effectively resolving the misclassification issue between open flames and lights in night-time construction scenarios, and providing precise and efficient intelligent technical support for fire risk prevention and control during the construction phase of nuclear power plants. Full article

Efficient vibration based walnut harvesting relies on the accurate determination of clamping-points on tree trunks. Properly selected clamping points can significantly enhance vibration transmission efficiency while minimizing mechanical damage to trees. However, most existing studies focus on developing generalized vibration models applicable to multiple trees, often overlooking the structural uniqueness of individual trees in clamping-point determination. This study proposes a monocular vision-based method for clamping-point determination in walnut vibration harvesting. Robustness and applicability in complex orchard environments are enhanced by introducing three keypoint annotation strategies with varying levels of structural constraints, namely 5-key-part annotation (5-KAS), 2-key-part annotation (2-KAS), and single-key-part annotation (1-KAS). Pose estimation models based on the YOLO architecture, including YOLOv8-pose, YOLO11-pose, and YOLOv12-pose, were evaluated to examine the effect of structure assisted annotation, and the results show that introducing structural constraints improves detection accuracy, training stability, and robustness. The YOLOv12-pose model combined with the 5-KAS achieves the best performance, with a precision of 95.8% and mean average precision (mAP) of 95.5%. Field harvesting experiments demonstrate that clamping-point prediction incorporating structural information achieves higher and more stable net harvesting rates. Overall, the proposed method offers a reliable and deployable solution for clamping-point determination using monocular RGB images, facilitating intelligent vibration harvesting in walnut orchards. Full article

In recirculating aquaculture systems, fish detection is an essential component for maintaining effective farming operations. The availability of high-quality fish datasets is limited because of the richness of fish species, and the annotation of large-scale data, which is used to train models, is often labor-intensive and time-consuming. The presence of different fish species across batches introduces further challenges for consistent detection performance. This work introduces a few-shot learning approach for fish detection, utilizing a customized dataset as novel classes and the Fish4Knowledge dataset for base classes, thereby establishing a framework that enhances adaptability in data-scarce scenarios. Within the model architecture, multi-scale feature extraction is enhanced through an attention mechanism, which is integrated as a dedicated module to strengthen representation learning, thus enhancing the model’s capability to differentiate visually similar fish species. Two distinct customized fish datasets are employed to evaluate the robustness of the proposed method. Experimental results show that the proposed model performs competitively against TFA, Meta-RCNN, and VFA. In the base-training phase, it achieves a mAP of 0.775, slightly surpassing VFA, while in the 1-shot, 5-shot, and 10-shot fine-tuning settings, it obtains mAP values of 0.152, 0.247, and 0.265, respectively. A similar trend is observed on a subset of black fish, with mAP scores of 0.169, 0.253, and 0.286 in the corresponding few-shot settings. These results indicate that the proposed approach can maintain relatively stable detection accuracy and adaptability across different fish batches, offering a practical solution for fish detection tasks in aquaculture when annotated data is scarce. To further demonstrate the efficacy and practical utility of the proposed methodology, a case study in fish farming confirms that the enhanced model achieves consistent and precise detection across diverse fish species, even when trained with limited annotated data. Full article

Autonomous driving technology is a core enabler for new energy vehicle industrial upgrading and a critical pillar for achieving sustainable development goals (SDGs), especially sustainable urban mobility, low-carbon transportation, and efficient intelligent transportation systems (ITS). However, unstable nighttime low-light perception severely restricts autonomous driving deployment, hindering sustainable transportation development—rooted in visual feature degradation and cross-modal imbalance that impair 3D object detection (autonomous driving’s core perception technology). To address this and advance sustainable autonomous driving, this paper proposes a Bird’s-Eye View (BEV)-based multi-modal 3D object detection approach tailored for nighttime scenarios, integrating low-light adaptive components while preserving the original BEV pipeline. Without modifying core inference, it enhances low-light robustness and cross-modal fusion stability, ensuring reliable perception for sustainable autonomous driving operation. Extensive experiments on the nuScenes nighttime subset quantify performance via rigorous metrics (NDS, mAP, mATE). Results show the method outperforms BEVFusion with negligible parameter/inference overhead, achieving 1.13% NDS improvement. This validates its effectiveness and provides a sustainable technical tool for autonomous driving perception, promoting new energy vehicle popularization, optimizing urban ITS efficiency, reducing perception-related accidents and carbon emissions, and directly contributing to transportation and socio-economic sustainability. Full article

Unmanned aerial vehicle (UAV) forest fire detection is vital for forest safety. However, early-stage UAV fire scenarios often involve small targets, weak smoke signals, and strict onboard resource constraints, which pose significant challenges to existing detectors. To improve the speed and accuracy of UAV forest fire detection, this paper proposes a lightweight fire detection algorithm, AHE-YOLO, specifically designed for UAVs. The proposed method adopts a coordinated lightweight design to improve feature preservation and cross-scale representation under limited computational budgets. Specifically, the Adaptive Downsampling (ADown) module preserves shallow fire-related cues during spatial reduction, improving sensitivity to small flame and smoke targets. The high-level screening-feature fusion pyramid network (HS-FPN) introduces cross-scale attention to promote more discriminative multi-level feature interaction while reducing redundant computation. Furthermore, the Efficient Mobile Inverted Bottleneck Convolution (EMBC) module is employed to improve receptive-field efficiency and feature selectivity under lightweight constraints, further enhancing detection accuracy and inference speed. Finally, the performance of AHE-YOLO is comprehensively evaluated through ablation and comparative experiments on the same dataset. The final experimental results show that YOLO-AHE achieves a mean average precision (mAP) of 94.8% while reducing model parameters by 39.7%, decreasing FLOPs by 27.0%, and shrinking the model size by 36.4%. In addition, its inference speed improves by 16.5%. Beyond detection performance, the proposed framework supports sustainable forest monitoring by enabling early fire warning with reduced computational and energy demands, showing strong potential for real-time deployment on resource-constrained UAV and edge platforms. Full article

Wafer defects in semiconductor manufacturing can directly damage the physical structure and circuit integrity of wafers, leading to the functional failure of chips. To address this problem, this paper proposes a Dual-Axis Attention-enhanced multi-scale fusion You Only Look Once version 13 (DAS-YOLOv13) model. Based on YOLOv13n, the model is specifically designed for the fast and accurate detection of tiny, multi-scale defects on wafer surfaces. It integrates innovative components such as a dual-axis attention module, an adaptive dynamic multi-scale representation module, and a self-modulation feature aggregation module. By enhancing salient feature expression, improving cross-scale representation capability, and optimizing deep semantic fusion strategies, the model achieves effective defect detection. On the wafer defect dataset, the DAS-YOLOv13 model achieves a mean Average Precision (mAP) of 74.2%, which is 4.3% higher than that of YOLOv13n; the Average Precision at an Intersection over Union (IoU) threshold of 50% (mAP50) reaches 92.9%. The results demonstrate that DAS-YOLOv13 effectively improves the detection accuracy of tiny, multi-scale defects through structural optimization. It provides a reliable solution for high-precision wafer detection in semiconductor manufacturing and can be seamlessly integrated into high-precision semiconductor automated inspection scenarios. Full article

To address the challenges of harvest crate localization caused by varying illumination, partial occlusion, and background interference in unstructured farmland environments, as well as the high costs and low efficiency associated with traditional manual harvesting, this paper proposes FAL-YOLO, a lightweight keypoint detection model. Using YOLOv8n-Pose as the baseline framework, the model integrates a C2f-ContextGuided backbone and a Slim-Neck feature fusion layer. Furthermore, a LSCD-LQE lightweight detection head is designed, and an Inner-MPDIoU loss function is introduced to enhance keypoint detection performance under complex backgrounds and occluded conditions. Experimental results on the self-constructed farmland harvest crate dataset indicate that FAL-YOLO requires only 1.71 M parameters and 4.5 GFLOPs of computational cost, representing reductions of 44.5% and 45.8% compared to YOLOv8n-Pose, while achieving an mAP@0.5 of 94.9%, corresponding to an improvement of 1.2%. Additionally, by establishing correspondences between keypoints and the 3D model through the PnP algorithm, the 3D pose of the crate can be reconstructed, providing reliable spatial input for robotic arm manipulation. The results demonstrate that FAL-YOLO achieves an effective balance between model lightweightness and detection accuracy, providing an efficient solution for automatic identification and grasping of harvest crates in farmland environments. Full article

Background/Objectives: Pressure injury stage classification was performed using object detection models to address inconsistencies in clinical assessment due to variability in nurses’ experience and education levels. Methods: A dataset of 1282 pressure injury images from a medical institution was used to train and compare five representative architectures, YOLOv5x, YOLOv7, YOLOv8x, YOLOv8n, and YOLOv11x, and Faster R-CNN across Stages 1–4, excluding Deep Tissue Injury and unclassified cases. A mobile application incorporating YOLOv7 was deployed at Eulji University Daejeon Medical Center and tested by 10 nurses over 2 weeks, processing 46 cases. Results: YOLOv7 demonstrated superior performance with mAP@0.5 of 0.97 and mAP@0.5:0.95 of 0.68, achieving 93% accuracy for Stage 2 classification, the most challenging diagnostic category. Clinical validation demonstrated 87% diagnostic accuracy, 4.0/5 user satisfaction, and workflow improvement with assessment time reduced from 4–6 min to 1 min. The application proved valuable as both a diagnostic support tool and educational resource for novice nurses, with zero critical misclassifications recorded. Conclusions: This study establishes the practical utility of AI-based pressure injury classification systems in clinical practice and their potential for enhancing nursing competency and workflow efficiency. Full article

3-Acetyl-1-propanol (3-AP) is a key intermediate in the pharmaceutical and pesticide industries, which can be synthesized from the biomass derivative 2-methylfuran (2-MF) through a one-step hydrogenation process with significant economic and environmental benefits. Zeolite-supported metal catalysts showed feasible application, but simply regulating the acidic sites was difficult to break the activity–selectivity balance. Traditional single-metal Pd-based catalysts still suffer from low dispersion. This study constructed the PdZn/TS-1 catalyst for the efficient conversion of 2-MF into 3-AP. The low electronegativity of Zn facilitates the electron transfer from Zn to Pd, forming an electron-rich Pd active center. A small amount of Zn embedded in the Pd lattice causes lattice contraction, optimizing the spatial configuration of active sites. The synergy between the electronic and structural effects significantly improves catalytic performance. Under optimized conditions, the conversion rate of 2-MF reached 80.6%, and the yield of 3-AP reached 69.1%, providing a new paradigm for the design of catalysts for the directed hydrogenation of furan derivatives. Full article

As deep learning models become increasingly integrated into critical decision-making systems, the need for xAI has grown paramount to ensure transparency, accountability, and trust.Post hocexplainability methods, which analyse trained models to interpret their predictions without modifying the underlying architecture, have become increasingly important, especially in fields such as healthcare and finance. Modern xAI techniques often produce feature importance rankings that fail to capture the true causal influence of features, particularly in transformer-based models. Recent quantitative metrics, such as Symmetric Relevance Gain (SRG), which measures the area between the feature corruption performance curves of the Most Important Feature (MIF) and the Least Important Feature (LIF), provide a more rigorous basis for evaluating explanation fidelity. In this study, we first show that existing xAI methods exhibit consistently poor performance under the SRG criterion when explaining transformer-based text classifiers. To address these limitations, we introduceEvoDropX, a novel framework that formulates explanation as an optimisation problem. EvoDropX leverages Grammatical Evolution (GE) to evolve sequences of feature corruption with the explicit objective of maximising SRG, thereby identifying features that most strongly influence model predictions. EvoDropX provides interventional, input–output (behavioural) explanations and does not attempt to infer or interpret internal model mechanisms. Through comprehensive experiments across multiple datasets (IMDB, Stanford Sentiment Treebank (SST-2), Amazon Polarity (AP)), multiple transformer models (BERT, roberta, distilbert), and multiple metrics (SRG, MIF, LIF, Counterfactual Conciseness (CFC)), we demonstrate that EvoDropX significantly outperforms all state-of-the-art (SOTA) xAI baselines including Attention-Aware Layer-Wise Relevance Propagation for Transformers (AttnLRP), SHapley Additive exPlanations (SHAP), and Local Interpretable Model-agnostic Explanations (LIME), when evaluated using intervention-based faithfulness criteria. Notably, EvoDropX achieves $74.77\%$ improvement in SRG than the best-performing baseline on the IMDB dataset with the BERT model, with consistent improvements observed across all dataset-model pairs. Finally, qualitative and linguistic analyses reveal that EvoDropX captures both sentiment-bearing terms and their structural relationships within sentences, yielding explanations that are both faithful and interpretable. Full article

As a next-generation immersive service, holographic video enables users to move freely within a virtual world. This imposes stringent requirements on wireless networks. Given the massive bandwidth capacity inherent to visible light, visible light communication (VLC) can effectively meet the transmission requirements of holographic video and is an ideal wireless technology for next-generation indoor immersive services. However, VLC channels are highly dependent on Line-of-Sight (LoS) links. Due to user mobility, traditional VLC systems relying on fixed-orientation Photodetectors (PDs) often suffer from severe channel fading, which significantly degrades the transmission performance. In this paper, we propose an indoor VLC holographic video transmission architecture supporting rotatable PDs, utilizing rotatable PDs mounted on Head-Mounted Displays (HMDs) to assist in holographic video transmission. To minimize the total transmission delay of all users, we address the holographic video transmission problem by jointly optimizing the transmit power allocation of VLC Access Points (APs) and the pitch and roll angles of the users’ PDs. By formulating the problem as a Markov Decision Process (MDP), we address it using a novel Deep Reinforcement Learning (DRL) strategy leveraging the Soft Actor–Critic (SAC) architecture. Simulation results demonstrate that the proposed scheme reduces the overall latency by up to 29.6% compared to the benchmark schemes. Furthermore, the convergence speed of the algorithm is improved by 35% compared to traditional deep reinforcement learning algorithms such as Deep Deterministic Policy Gradient (DDPG). Full article

In complex traffic environments, image degradation caused by haze, low illumination, and occlusion significantly undermines the reliability of vehicle and pedestrian detection. To address these challenges, this paper proposes an aerial vision framework that tightly couples multi-level image enhancement with a lightweight detection architecture. At the image preprocessing stage, a cascaded “dehazing + enhancement” module is constructed, where a learning-based dehazing method is employed to restore long-range details affected by scattering artifacts. Additionally, structural fidelity is enhanced in low-light regions, while global brightness consistency is achieved. On the detection side, a lightweight yet robust detection architecture, termed GDEIM-SF, is designed. It adopts GoldYOLO as the lightweight backbone and integrates D-FINE as an anchor-free decoder. Moreover, two key modules, CAPR and ASF, are incorporated to enhance high-frequency edge modeling and multi-scale semantic alignment. Through evaluation on the VisDrone dataset, the proposed method achieves improvements of approximately 2.5 to 2.7 percentage points in core metrics such as mAP@50-90 compared to similar lightweight models, while maintaining a low parameter count and computational overhead. This ensures a balanced trade-off among detection accuracy, inference efficiency, and deployment adaptability, providing a practical and efficient solution for UAV-based visual perception tasks under challenging imaging conditions. Full article

Wind turbines operating in harsh environments are prone to surface defects that compromise efficiency and safety. Traditional convolutional neural networks lack sufficient multi-scale feature representation, while Transformer-based methods suffer from excessive computational complexity. This study proposes HAMD-DETR, an end-to-end detection framework for wind turbine defect identification. The framework consists of three key components: an Adaptive Dynamic Multi-scale Perception Network (ADMPNet), a Hierarchical Dynamic Feature Pyramid Network (HDFPN), and a Dynamic Frequency-Domain Feature Encoder (DFDEncoder). Firstly, ADMPNet integrates multi-scale dynamic integration fusion and adaptive inception depthwise convolution for feature extraction. Then the HDFPN balances deep semantic and shallow detail features through pyramid adaptive context extraction and gradient refinement modules. At last, DFDEncoder enhances feature discrimination through frequency-domain transformation. Experiments on wind turbine datasets demonstrate that HAMD-DETR achieves 58.6% mAP50 and 31.7% mAP50-95, representing improvements of 3.1% and 2.1% over the baseline RT-DETR. The proposed method reduces computational complexity by 27.2% and parameters by 30% while achieving a 151.9 FPS inference speed. These results validate HAMD-DETR’s effectiveness for wind turbine defect detection and demonstrate its potential for intelligent operation and maintenance applications. Full article

The dehydration response element binding protein (DREB) family of the AP2/ERF superfamily functions as a key regulatory component in plant adaptation to water-deficit conditions. However, studies on the DREB A4 subfamily in poplar (Populus trichocarpa) are insufficient. In this study, members of the DREB A4 subgroup in poplar were identified and analyzed via bioinformatic analysis. A pCAMBIA-2300-PtrDREB4 expression vector was constructed and transformed into Arabidopsis, followed by phenotypic analysis of transgenic plant in response to drought stress. A total number of 29 DREB A4 members were identified in the poplar genome. Synteny analysis revealed that 19 gene pairs went through segmental duplication at least 12.84 million years ago. Their promoter regions were enriched with cis-elements related to stress resistance, hormone regulation, and growth and development. Upstream regulator analysis of poplar DREB A4 genes identified 425 transcription factor genes, which belonged to 39 families. Gene expression analysis demonstrated distinct expression patterns of DREB A4 genes in leaves, roots and stems with a notable response to drought stress. Ectopic expression of PtrDREB4 in yeast and Arabidopsis increased the drought tolerance of transformants, indicating the positive role of PtrDREB4 in response to drought stress. These findings collectively established a theoretical foundation for further functional exploration of DREB A4 genes in poplar. Full article

People affected by vision impairment experience significant challenges in mobility and daily life activities. In this paper, a smart assistive navigation system is proposed to address mobility challenges and to enhance the independence of visually impaired individuals. Three modules are integrated into the proposed system. The vision module detects obstacles and interactive objects such as doors, chairs, people, fire extinguishers, etc. The depth cam-based distance module provides the distance of detected objects and obstacles. The voice module provides auditory feedback to visually impaired individuals about the detected objects and obstacles that fall under the pre-defined threshold distance. Finally, the proposed system is optimized in terms of performance and user experience. Jetson Nano is used to reduce the cost of the overall system; however, it has compatibility issues with many of the latest object detection models. The YOLOv5n model is used considering compatibility for object detection, but it has low Mean Average Precision (mAP) and frame rate. To improve the performance of the vision module, various hyperparameters of YOLOv5n are fine-tuned along with transfer learning to enhance the mAP@50 from the original 0.457 to 0.845 and mAP@50-95 from 0.28 to 0.593. Tensor-RT optimization is employed to increase the frame rate to deploy the model in a real scenario. The real-time experimentation shows that the proposed system successfully alerts users to key objects, hazards, and obstacles which enables independent and confident navigation. Full article