Search Results (707)

Aircraft cargo compartment fires constitute a significant type of aviation fire, posing a grave threat to aviation safety. To guard against and respond to such fires, existing aircraft cargo compartments are equipped with smoke detection fire detectors, which rely on perceiving changes in smoke transmittance to determine the onset of a fire. However, these detectors offer relatively low recognition accuracy and cannot provide a direct visual representation of the fire. In this work, we introduce a fire recognition method built on image sensors and a deep learning model. In light of the irregular shapes of flames and smoke, an improved interactive triplet attention mechanism (ITAM) is integrated into the You Only Look Once version 5 (YOLOv5) model, enhancing the model’s recognition accuracy. Furthermore, the original Neck structure is replaced with an Asymptotic Feature Pyramid Network (AFPN), improving the model’s ability to recognize small targets, which is particularly useful for detecting flames and smoke early in a fire. This paper further improves the model’s recognition accuracy by introducing the Focaler-IoU loss function, which balances the feature learning of hard and easy samples. Therefore, the network model in this paper is named IATF-YOLO. Ablation experiments demonstrate that our algorithm improves accuracy by 2%, while comparative experiments with several mainstream baseline models show that our algorithm achieves a 0.7% accuracy improvement, with a final peak accuracy of 93.6%. Full article

The You Only Look Once (YOLO) series has consistently advanced the field of object detection, evolving from YOLOv1 to the latest YOLO26, achieving remarkable improvements in detection accuracy and computational efficiency. However, deploying such high-performance models on resource-constrained edge devices remains challenging, particularly for tasks requiring real-time inference. A critical yet often overlooked factor affecting edge deployment is the trade-off between input image resolution and computational cost: while higher resolution preserves fine-grained details essential for detecting small defects, it proportionally increases energy consumption and latency. To address this issue, we propose a novel multi-resolution adaptive detection framework based on YOLO26, specifically optimized for Ascend NPU edge devices. Our method dynamically selects the most suitable input resolution for each inference instance via a jointly optimized scene complexity metric, where the feature weights and resolution thresholds are simultaneously calibrated through Bayesian multi-objective optimization to achieve an optimal balance between predictive accuracy and energy efficiency. The experiments on transmission line insulator defect detection demonstrate that our approach achieves favorable trade-offs, maintaining high detection precision while significantly reducing power consumption compared to fixed-resolution baselines. The proposed framework provides a viable solution for intelligent visual inspection in power grid infrastructure. Full article

To mitigate the drawbacks of joint crossover (IoU) in complex detection scenarios, this paper proposes a normalized IoU strategy. This strategy enhances the matching robustness in multi-scale object detection by introducing target scale parameters. The proposed method shows comparable or superior average precision (mAP) performance to traditional methods on public datasets. In addition, we have designed a dual-center distance penalty mechanism that implicitly enforces symmetric constraints between bounding boxes, increasing the number of positive samples detected. Our method has been evaluated on mainstream public datasets and unmanned aerial vehicle (UAV) water level gauge datasets, as well as evaluated using the You Only Look Once (YOLO) framework. Our method increased the average number of positive samples by 2.28% compared to CIoU. It also surpasses the most advanced technology. The dual-center constraint enhances the spatial alignment between bounding boxes. This results in notable performance gains in challenging scenarios. These scenarios involve blurred and heavily occluded objects. After parameter optimization, the proposed method achieves significant accuracy improvements. These improvements are seen in detecting small-scale and occluded characters. Full article

Birds, as essential components of biodiversity, serve as critical indicators of ecosystem health and stability through their population dynamics and spatial distribution. However, the complexity of natural habitats, occlusion caused by avian behavior, and logistical challenges in field monitoring pose significant difficulties for fine-grained bird detection. To address these issues, this paper presents NEB-YOLO, a lightweight bird detection network built upon an improved YOLOv8 (You Only Look Once version 8) architecture. First, to enhance detection capability for fine-grained bird images in complex backgrounds, an Efficient Multi-scale Attention (EMA) mechanism is integrated into the Neck network of the YOLOv8n architecture to construct the teacher network. Second, to accommodate resource constraints in practical scenarios, channel-wise Group_Slim pruning is applied to reduce the parameter count and computational overhead of the student model. Third, to achieve high accuracy with lightweight models, a joint offline distillation approach is adopted that combines logic-based knowledge distillation (BCKD) with feature-based knowledge distillation (CWD). This design facilitates effective transfer of discriminative features for bird imagery from the teacher model to the student model. Experiments demonstrate that the proposed framework achieves a performance gain of over 1.3% in $mAP$ on the fine-grained CUB-200-2011 bird dataset, while reducing model size by 67% compared to the original model. These results confirm that NEB-YOLO strikes an effective balance between model size and accuracy, indicating its suitability for resource-constrained scenarios. Full article

Affected by global climate change and complex environmental factors, the frequency and intensity of forest fires have been rising. Accurate early detection is crucial for disaster mitigation. Traditional methods (e.g., manual monitoring) suffer from low efficiency or limited coverage, while deep learning methods (e.g., YOLO (You Only Look Once), Faster RCNN (Region-based Convolutional Neural Networks)) perform well but are sensitive to degraded images (haze, low light), reducing accuracy. To address blurred smoke features and attenuated flame brightness in degraded images, this paper proposes CoDeF-Net (Collaborative Detection Framework Network), a collaborative detection framework integrating Retinex-BCE (Retinex-based Bright Channel Enhancement) image enhancement with YOLOv11 (You Only Look Once version 11) to improve robustness. Experiments on 1757 real forest fire images show that Retinex-BCE achieves an FSIMC (Full-Reference Image Quality Assessment Metric based on Structural Similarity and Contrast) index of 0.9611 and an LOE (Loss of Edge) value of 254.78, preserving image structure. CoDeF-Net reaches AP@0.5 (Average Precision at Intersection over Union threshold 0.5) of 87.9% (3.8% higher than original YOLOv11), with low missed detection of small flames and enhanced stability in extreme scenarios, providing a feasible solution for forest fire monitoring under degraded images. Full article

To address critical challenges in maritime ship detection within complex surveillance imagery, including severe background interference, extreme scale variation, and fine-grained category confusion, this study proposes Maritime Scene Collaborative You Only Look Once (MSC-YOLO), an improved detection model for fixed-location maritime surveillance scenarios. First, a Maritime Scene Adaptive Attention Module (MSAM) is introduced to suppress water-surface clutter and enhance structurally informative ship responses through bidirectional feature regulation, thereby strengthening feature representation in background-complex scenes. In addition, a Scale-aware Dynamic Head (SDA-Head) is designed by integrating deformable convolution with parallel scale-aware prediction branches to improve detection coverage for vessels under pronounced scale variation. Furthermore, a Class Prototype Guided (CPG) module is developed, incorporating class-level prototypes and category-similarity priors to improve the discriminative representation of visually similar ship categories and component states. Experimental results on the constructed maritime surveillance dataset show that MSC-YOLO achieves 0.9723 mAP@50, 0.7315 mAP@50–95, 0.8903 Precision, and 0.9883 Recall. Compared with YOLOv11n, the proposed model improves mAP@50 by 17.77%, Precision by 21.82%, and Recall by 8.16%, indicating clear advantages in target discovery, clutter robustness, and difficult-target coverage in complex maritime surveillance scenes. Visualization and confusion-matrix analyses further show reduced background interference and stronger class-wise discrimination. Overall, MSC-YOLO demonstrates effective and reliable performance for complex maritime surveillance scenarios. Full article

In terms of the detection of colorectal polyps during a colonoscopy, the accuracy of the diagnosis is key to effective prevention and treatment, and can be hindered by manual identification. Colorectal polyps are abnormal tissue growths in the colon or rectum, and their sizes, shapes and textures can make them difficult to find. Researchers have now turned to deep learning techniques and the YOLOv11 detection framework in particular to provide a method to automate the recognition and accurate identification of these abnormal growths. Specifically, the proposed method modifies the conventional YOLOv11 detection workflow by generating bounding box annotations from polyp segmentation masks, applying region-aware data preprocessing and augmentation, and training the detector under region-guided supervision to enhance localization precision and detection robustness. polyp segmentation masks are utilized to generate bounding box annotations which not only contribute exact spatial supervision but also avoid manual box labeling inconstancy. Region-aware data preprocessing and augmentation pay more attention to polyp-relevant regions and suppress background noise, which leads to clearer feature discrimination for small or irregular polyps. Additionally, region-guided supervision serves as explicit guidance for localizing objects with the anatomical polyp regions, which largely helps achieve accurate boundaries and prevent false detections. The proposed YOLOv11-based polyp detection system was tested and evaluated on the publicly available Kvasir-SEG dataset, which is comprised of annotated colonoscopy images. Enhanced data pre-processing and exhaustive training with appropriate choice of hyper-parameters fortified the reliability and useability of the model. The results confirmed high-grade results, and gave an Intersection over Union score of 0.9764, and an overall correctness rate of 99.00%, with well-balanced precision, recollection and F1-scores. Coming in with a mean Average Precision (mAP) of 0.9937 at a Intersection over Union threshold of 0.5 and 0.9935 over the full spectrum of thresholds from 0.5 to 0.95, this shows that the model is able to consistently and reliably detect polyps. The proposed system was also compared with Segment Anything Model, YOLO-Seg, and SAM2 and confirmed the efficacy of its method. Full article

Wind turbine blade defect detection requires accurate identification of small and irregular defects while maintaining low computational cost for practical inspection scenarios. However, lightweight detectors often suffer from insufficient local feature extraction, limited multiscale feature fusion, and weak responses to critical defect regions. To address these issues, this study proposes a Receptive-Field-Enhanced You Only Look Once model (RFE-YOLO), a lightweight defect detection model based on You Only Look Once version 10 nano (YOLOv10n).The proposed model introduces three task-oriented improvements. First, C2f-RFAConv is embedded into the backbone to enhance receptive field aware local feature representation for fine grained defects. Second, a Compact Cross-scale Feature Fusion Module, termed CCFM, is designed in the neck to improve the integration of low-level detail information and high-level semantic features with reduced computational complexity. Third, an Efficient Local Attention module is inserted before the detection head to strengthen defect-related spatial responses after feature fusion. Experiments were conducted on a wind turbine blade defect dataset containing three categories, namely Crack, Oil leakage, and Peel. The results show that RFE-YOLO achieves 89.9% mean Average Precision at an Intersection over Union threshold of 0.5, namely mAP@0.5, and 64.73% mAP@0.5:0.95. Compared with YOLOv10n, RFE-YOLO improves mAP@0.5 by 2.8 percentage points while reducing the number of parameters from 2.70M to 1.91M and giga floating point operations from 8.4 to 5.3. The inference speed reaches 88.8 frames per second on an NVIDIA GeForce RTX 3090 GPU. These results indicate that RFE-YOLO achieves a favorable balance between detection accuracy and model efficiency under the current experimental setting. Full article

Monitoring marine organisms plays a vital role in biodiversity conservation, marine environmental management, and fisheries resource management. However, the underwater environment is often low-light and turbid, leading to indistinct target boundaries. Moreover, the wide variety of marine organisms—with significant differences in color, scale, texture, and morphology—can easily result in missed detections. To address these challenges, this paper proposes a multi-class marine organism detection method using multi-scale attention-enhanced You Only Look Once 11 nano (YOLO11n). The method incorporates the Convolutional Block Attention Module (CBAM) into the YOLO11n network, enabling the model to better focus on key feature regions while effectively suppressing background noise interference in complex marine environments. In addition, the model is trained using the Complete Intersection over Union (CIoU) loss function, which enhances bounding box regression accuracy, especially in handling targets of varying scales. The effectiveness of the proposed method is validated on the publicly available BrackishMOT dataset. The proposed model achieves an overall mAP@0.5 of 0.481, computed as the average AP across six organism categories. Category-wise results indicate stronger performance on visually distinguishable targets, such as Jellyfish, Starfish, and Small fish, with AP values of 0.808, 0.678, and 0.677, respectively. In contrast, performance remains limited for rare or visually ambiguous categories. These results suggest that the proposed method is effective for multi-class marine organism detection, particularly when discriminative visual features are present. Full article

Efficient detection of fish diseases is essential for intelligent health monitoring and timely intervention in aquaculture. However, current computer vision models remain computationally intensive, hindering their deployment on resource-constrained edge devices in aquaculture applications. To this end, this study developed a lightweight detection framework based on an improved You Only Look Once (YOLO), aiming to achieve a favorable balance between detection accuracy and on-site inference efficiency. First, a Dual-Branch Feature-Preserving Downsampling (DFPD) module was proposed to enhance the extraction of valuable disease-related cues with minimal computational overhead. Subsequently, structured pruning was applied to compress the optimized baseline model. Four pruning techniques, including Slim, GroupTaylor, Layer-Adaptive Magnitude-Based Pruning (LAMP), and L1-based, were evaluated under the same conditions. The enhanced baseline model improved precision from 0.864 to 0.908 and mAP@0.5:0.95 from 0.613 to 0.632, while already reducing the Number of Parameters (Params) and Giga Floating-point Operations Per Second (GFLOPs) compared with the original YOLOv8n. Among the pruning techniques, L1-based produced the best overall trade-off, yielding a final model that maintained a F1-score of 0.860 while reducing Params and GFLOPs by 54.7% and 49.4%, respectively, relative to the original detector. Ablation studies further revealed that a moderate FLOPs reduction of approximately 41% to 47% was optimal for preserving diagnostic performance while enhancing compactness. Edge deployment tests on an RK3588S device verified the framework’s practical inference speed advantage. Therefore, this study offers a deployment-friendly computer vision solution for on-site fish disease detection in aquaculture management, particularly suited to real-world scenarios with limited computational resources. Full article

Small object detection in Unmanned Aerial Vehicle (UAV) imagery faces critical challenges, including extreme scale variations, dense spatial distributions, and stringent computational constraints, for real-time deployment. To address these challenges, this paper proposes a CBW-based Detection Transformer (CBW-DETR), an enhanced transformer-based detection framework that integrates architectural efficiency with scale-aware mechanisms throughout the detection pipeline. The framework comprises three coordinated innovations. First, a Context-Guided Feature Extraction (ContextGFE) module reduces model parameters and theoretical computational cost through adaptive receptive field selection and wavelet-domain enhancement while maintaining representational capacity. Second, a Scale-Aware Feature Pyramid Network (SAFPN) employs spatial-variant compensation factors and cross-scale attention to facilitate balanced gradient flow across pyramid levels, particularly benefiting small object detection. Third, an Adaptive Scale IoU (ASIoU) loss function implements uncertainty-aware gradient modulation and scale-specific optimization to enhance localization accuracy for objects of varying sizes. Extensive experiments on VisDrone2019 and Dataset for Object Detection in Aerial Images (DOTA) datasets demonstrate that CBW-DETR achieves substantial improvements in detection accuracy while reducing model parameters by 28.1% and theoretical computation by 18.0% compared to the Real-Time Detection Transformer-R18 (RT-DETR-R18) baseline. These reductions in model complexity come at a moderate cost in inference throughput (73.6 frames per second (FPS) vs. 94.1 FPS), attributable to memory-access-intensive operations introduced by multi-branch convolutions and wavelet transforms. Among the evaluated detectors, including You Only Look Once (YOLO) series variants and transformer-based methods, CBW-DETR achieves a competitive detection accuracy with a notably compact model footprint. Visualization analysis confirms its robust performance across diverse challenging scenarios including nighttime conditions, dense object distributions, and severe occlusions, validating the framework’s practical applicability for UAV-based detection applications. Full article

With the development of the Internet of Things (IoT) and microelectronics technology, the methods used to tamper with fuel dispensers have become increasingly concealed, posing significant challenges to market supervision and law enforcement. This paper releases a tampering features dataset of assembled printed circuit boards (PCBs) from fuel dispensers, aiming to provide high-quality data support for automated, computer-vision-based illicit metering detection. The dataset encompasses multi-class tampering features derived from 189 high-resolution images of PCBs seized during real-world law enforcement, covering 5 mainstream brands. To eliminate perspective bias, rigorous lens distortion correction and four-point homography transformation preprocessing were conducted on the images. Additionally, six typical tampering features (e.g., the addition of tampered surface-mount resistors) were manually and precisely annotated, and then cross-checked and confirmed by domain experts. Furthermore, the dataset was benchmarked using multiple generations of You Only Look Once (YOLO) object detection models (Baseline Validation), which have been demonstrated to handle both large and small object detection in high-resolution images. The evaluation results, including confusion matrices and t-distributed Stochastic Neighbor Embedding (t-SNE) feature clustering diagrams, demonstrate the reliability and effectiveness of this dataset for training high-precision fraud detection models. This dataset is intended to support computer vision and anti-fraud research, promoting the automated development of fuel dispenser tampering detection. Full article

Accurate fruit ripeness assessment is essential for optimizing harvest timing and ensuring postharvest quality, particularly in climacteric fruits such as peaches, which exhibit rapid and heterogeneous ripening. This study proposes a deep learning-based approach for multistage peach ripeness classification across five maturity levels (green, green-blushed, blushed, yellow-blushed, and fully yellow). Four datasets were constructed using controlled image acquisition, segmentation, data augmentation, and perceptual hashing to mitigate data leakage. The performance of AlexNet, EfficientNet-B0, and three YOLO (You Only Look Once) architectures (YOLOv8, YOLOv11, and YOLOv12) was evaluated using standard metrics, including accuracy, precision, recall, F1 score, mAP, and inference speed. Results show that YOLO-based models significantly outperform classical networks, achieving accuracies between 95.25% and 98.3% and mAP@0.5 above 98.25%, while also reducing inference time to 8.1–12.7 ms compared with 722.23 ms for AlexNet and 171.87 ms for EfficientNet-B0. In a practical sorting experiment with 214 peaches, YOLOv12 achieved 92.06% accuracy, demonstrating robust real-world performance. Misclassifications were primarily observed between adjacent ripeness stages. These findings indicate that YOLO-based models provide an effective and scalable solution for real-time fruit sorting, while the use of perceptual hashing enhances dataset reliability and model generalization for deployment in agricultural quality control systems. Full article

Transient earth voltage (TEV) measurements are widely used for partial discharge (PD) monitoring in medium-voltage equipment due to their non-intrusive nature and suitability for field deployment. However, TEV-based PD analytics remain challenging in practical environments because PD signatures often overlap with noise characteristics. Recent advances in deep learning enable the image-based analysis of phase-resolved partial discharge (PRPD) representations, offering improved robustness compared to conventional signal-based methods. This paper presents a structured experimental investigation of classification-based and localisation-based deep learning models for TEV PRPD analytics. A multi-column convolutional neural network (MCNN) is evaluated as a classification model, while YOLOv5 and YOLOv12, based on the You Only Look Once (YOLO) framework, are investigated as object detection frameworks capable of PD localisation. Experiments are conducted using both original and expanded datasets, with additional analysis on the impact of hyperparameter optimisation. The results show that dataset expansion and hyperparameter tuning improve classification and detection performance for several models. The MCNN achieves strong classification accuracy for PD/noise screening, while YOLOv5 demonstrates substantial improvement in localisation performance after optimisation. In contrast, YOLOv12 maintains stable detection performance under the evaluated training configurations. The results highlight the trade-offs between classification accuracy, localisation capability, and annotation complexity when selecting deep learning models for PD analytics. These findings provide practical insights into the deployment of deep learning techniques for TEV-based PD monitoring systems and highlight the complementary roles of classification and localisation frameworks in PRPD analysis. Full article

An AI video–based assessment system is used to analyze the volleyball forearm pass under continuous wall-volley conditions in this study. A single 120 frames per second (FPS) high-speed camera captures the athlete from a rear-oblique view. A laptop executes a You Only Look Once (YOLO)-based pipeline to detect the ball and human keypoints, including the shoulders, elbows, wrists, hips, knees, and ankles. From the joint angles and ball–body relative positions, three cues are quantified. The first cue is the ready posture, characterized by straight arms, downward wrist flexion, an upper arm–trunk angle of approximately 90°, and a forward-leaning center of mass. The second cue is the ball–contact point located posterior to the wrist joint. The third cue is the variation in the center of mass synchronized with the rhythm of the ball. Five athletes performed ten trials, and the predictions were compared against manual annotations, achieving greater than 95% accuracy in criterion attainment. The system outputs criterion scores and key frames to provide immediate feedback. Deployment challenges, including occlusion, viewpoint, and illumination, are discussed, along with potential extensions such as multi-camera fusion and temporal tracking. Full article