Search Results (3,843)

Given the high computational complexity and poor real-time performance of current photovoltaic cell surface defect detection methods, this study proposes a lightweight model, YOLOv8-FSD, based on YOLOv8. By introducing the FasterNet network to replace the original backbone network, computational complexity and memory access are reduced. A thin neck structure designed based on hybrid convolution technology is adopted to reduce model parameters and computational load further. A lightweight dynamic feature upsampling operator improves the feature map quality. Additionally, the regularized Gaussian distribution distance loss function is used to enhance the detection ability for small target defects. Experimental results show that the YOLOv8-FSD lightweight algorithm improves detection accuracy while significantly reducing the number of parameters and computational requirements compared to the original algorithm. This improvement provides an efficient, accurate, and lightweight solution for PV cell defect detection. Full article

Oudemansiella raphanipes is valued for its rich nutritional content and medicinal properties, but traditional manual grading methods are time-consuming and labor-intensive. To address this, deep learning techniques are employed to automate the grading process, and knowledge distillation (KD) is used to enhance the accuracy of a small-parameter model while maintaining a low resource occupation and fast response speed in resource-limited devices. This study employs a three-teacher KD framework and investigates three cascaded structures: the parallel model, the standard series model, and the series model with residual connections (residual-series model). The student model used is a lightweight ShuffleNet V2 0.5x, while the teacher models are VGG16, ResNet50, and Xception. Our experiments show that the cascaded structures result in improved performance indices, compared with the traditional ensemble model with equal weights; in particular, the residual-series model outperforms the other models, achieving a grading accuracy of 99.7% on the testing dataset with an average inference time of 5.51 ms. The findings of this study have the potential for broader application of KD in resource-limited environments for automated quality grading. Full article

Powder bed fusion–laser beam (PBF-LB) additive manufacturing enables the production of intricate, lightweight metal components aligned with Industry 4.0 and sustainable principles. However, residual stresses and distortions challenge the dimensional accuracy and reliability of parts. Inherent strain methods (ISMs) provide a computationally efficient approach to predicting these issues but often overlook transient thermal histories, limiting their accuracy. This paper introduces an enhanced inherent strain method (EISM) for PBF-LB, integrating macro-scale temperature histories into the inherent strain framework. By incorporating temperature-dependent adjustments to the precomputed inherent strain tensor, EISM improves the prediction of residual stresses and distortions, addressing the limitations of the original ISM. Validation was conducted on two Ti-6Al-4V geometries—a non-symmetric bridge and a complex structure (steady blowing actuator)—through comparisons with experimental measurements of temperature, distortion, and residual stress. Results demonstrate improved accuracy, particularly in capturing localized thermal and mechanical effects. Sensitivity analyses emphasize the need for adaptive layer lumping and mesh refinement in regions with abrupt stiffness changes, such as shrink lines. While EISM slightly increases computational cost, it remains feasible for industrial-scale applications. This work bridges the gap between simplified inherent strain models and high-fidelity simulations, offering a robust tool for simulation-driven optimisation. Full article

Large vision language models (LVLMs) are built upon large language models (LLMs) and incorporate non-textual modalities; they can perform various multimodal tasks. Applying LVLMs in remote sensing (RS) visual question answering (VQA) tasks can take advantage of the powerful capabilities to promote the development of VQA in RS. However, due to the greater complexity of remote sensing images compared to natural images, general-domain LVLMs tend to perform poorly in RS scenarios and are prone to hallucination phenomena. Multi-agent debate for collaborative reasoning is commonly utilized to mitigate hallucination phenomena. Although this method is effective, it comes with a significant computational burden (e.g., high CPU/GPU demands and slow inference speed). To address these limitations, we propose Co-LLaVA, a model specifically designed for RS VQA tasks. Specifically, Co-LLaVA employs model collaboration between Large Language and Vision Assistant (LLaVA-v1.5) and Contrastive Captioners (CoCas). It combines LVLM with a lightweight generative model, reducing computational burden compared to multi-agent debate. Additionally, through high-dimensional multi-scale features and higher-resolution images, Co-LLaVA can enhance the perception of details in RS images. Experimental results demonstrate the significant performance improvements of our Co-LLaVA over existing LVLMs (e.g., Geochat, RSGPT) on multiple metrics of four RS VQA datasets (e.g., +3% over SkySenseGPT on “Rural/Urban” accuracy in the test set of RSVQA-LR dataset). Full article

Multi-class object detectors often suffer from the class imbalance issue, where substantial model performance discrepancies exist between classes. Generative adversarial networks (GANs), an emerging deep learning research topic, are able to learn from existing data distributions and generate similar synthetic data, which might serve as valid training data for improving object detectors. The current study investigated the utility of lightweight unconditional GAN in addressing weak object detector class performance by incorporating synthetic data into real data for model retraining, under an agricultural context. AriAplBud, a multi-growth stage aerial apple flower bud dataset was deployed in the study. A baseline YOLO11n detector was first developed based on training, validation, and test datasets derived from AriAplBud. Six FastGAN models were developed based on dedicated subsets of the same YOLO training and validation datasets for different apple flower bud growth stages. Positive sample rates and average instance number per image of synthetic data generated by each of the FastGAN models were investigated based on 1000 synthetic images and the baseline detector at various confidence thresholds. In total, 13 new YOLO11n detectors were retrained specifically for the two weak growth stages, tip and half-inch green, by including synthetic data in training datasets to increase total instance number to 1000, 2000, 4000, and 8000, respectively, pseudo-labeled by the baseline detector. FastGAN showed its resilience in successfully generating positive samples, despite apple flower bud instances being generally small and randomly distributed in the images. Positive sample rates of the synthetic datasets were negatively correlated with the detector confidence thresholds as expected, which ranged from 0 to 1. Higher overall positive sample rates were observed for the growth stages with higher detector performance. The synthetic images generally contained fewer detector-detectable instances per image than the corresponding real training images. The best achieved YOLO11n AP improvements in the retrained detectors for tip and half-inch green were 30.13% and 14.02% respectively, while the best achieved YOLO11n mAP improvement was 2.83%. However, the relationship between synthetic training instance quantity and detector class performances had yet to be determined. GAN was concluded to be beneficial in retraining object detectors and improving their performances. Further studies are still in need to investigate the influence of synthetic training data quantity and quality on retrained object detector performance. Full article

Unmanned Aerial Vehicle (UAV) object detection is crucial in various fields, such as maritime rescue and disaster investigation. However, due to small objects and the limitations of UAVs’ hardware and computing power, detection accuracy and computational overhead are the bottleneck issues of UAV object detection. To address these issues, a novel convolutional neural network (CNN) model, LCSC-UAVNet, is proposed, which substantially enhances the detection accuracy and saves computing resources. To address the issues of low parameter utilization and insufficient detail capture, we designed the Lightweight Shared Difference Convolution Detection Head (LSDCH). It combines shared convolution layers with various differential convolution to enhance the detail capture ability for small objects. Secondly, a lightweight CScConv module was designed and integrated to enhance detection speed while reducing the number of parameters and computational cost. Additionally, a lightweight Contextual Global Module (CGM) was designed to extract global contextual information from the sea surface and features of small objects in maritime environments, thus reducing the false negative rate for small objects. Lastly, we employed the WIoUv2 loss function to address the sample imbalance issue of the datasets, enhancing the detection capability. To evaluate the performance of the proposed algorithm, experiments were performed across three commonly used datasets: SeaDroneSee, AFO, and MOBdrone. Compared with the state-of-the-art algorithms, the proposed model showcases improvements in mAP, recall, efficiency, where the mAP increased by over 10%. Furthermore, it utilizes only 5.6 M parameters and 16.3 G floating-point operations, outperforming state-of-the-art models such as YOLOv10 and RT-DETR. Full article

The widespread deployment of CCTV systems has significantly enhanced surveillance and public safety across various environments. However, the emergence of deepfake technology poses serious challenges by enabling malicious manipulation of video footage, compromising the reliability of CCTV systems for evidence collection and privacy protection. Existing deepfake detection solutions often suffer from high computational overhead and are unsuitable for real-time deployment on resource-constrained CCTV cameras. This paper proposes FL-TENB4, a Federated-Learning-enhanced Tiny EfficientNetB4-Lite framework for deepfake detection in CCTV environments. The proposed architecture integrates Tiny Machine Learning (TinyML) techniques with EfficientNetB4-Lite, a lightweight convolutional neural network optimized for edge devices, and employs a Federated Learning (FL) approach for collaborative model updates. The TinyML-based local model ensures real-time deepfake detection with minimal latency, while FL enables privacy-preserving training by aggregating model updates without transferring sensitive video data to centralized servers. The effectiveness of the proposed system is validated using the FaceForensics++ dataset under resource-constrained conditions. Experimental results demonstrate that FL-TENB4 achieves high detection accuracy, reduced model size, and low inference latency, making it highly suitable for real-world CCTV environments. Full article

This study optimizes the structural design of a composite wing shell by minimizing mass and maximizing the first natural frequency. The analysis focuses on the effects of polyvinyl chloride (PVC) foam thickness and the fiber orientation angle of the inner carbon layers, with the outer layers fixed at ±45° for torsional rigidity. A Multi-Objective Genetic Algorithm (MOGA), well suited for complex engineering problems, was employed alongside Design of Experiments to develop a precise response surface model, achieving predictive errors of 0% for mass and 2.99% for frequency. The optimal configuration—90° and 0° fiber orientations for the upper and lower layers and a foam thickness of 1.05 mm—yielded a mass of 412 g and a frequency of 122.95 Hz. These findings demonstrate the efficacy of MOGA in achieving innovative lightweight aerospace designs, striking a balance between material efficiency and structural performance. Full article

Face detection in resource-constrained environments presents challenges, due to the computational demands of state-of-the-art models and the complexity of real-world conditions, such as variations in scale, pose, and occlusion. This study introduces FeatherFace, a lightweight face-detection architecture with only 0.49 M parameters, designed for high accuracy and efficiency in such environments. Leveraging MobileNet-0.25 as a backbone, FeatherFace incorporates advanced feature-integration strategies, including a bidirectional feature pyramid network (BiFPN), a convolutional block attention module (CBAM), deformable convolutions, and channel shuffling. Evaluated on the WIDERFace dataset, FeatherFace achieves an overall average precision (AP) of 87.2%, with notable performance gains of 4.0% AP on the Hard subset compared with the baseline. Ablation studies highlight the critical role of multiscale feature integration and the strategic placement of attention mechanisms in addressing detection challenges such as small or occluded faces. With its compact design and reduced inference time, FeatherFace bridges the gap between the reliability of computationally intensive models and the need for deploying robust models in highly resource-constrained environments, such as edge devices and embedded systems. This work provides valuable insights for developing robust and lightweight models suited to challenging real-world applications. Full article

In view of the problems that mean that existing detection networks are not effective in detecting dynamic targets such as wildfire smoke, a lightweight dynamically enhanced transmission line channel wildfire smoke detection network LDENet is proposed. Firstly, a Dynamic Lightweight Conv Module (DLCM) is devised within the backbone network of YOLOv8 to enhance the perception of flames and smoke through dynamic convolution. Then, the Ghost Module is used to lightweight the model. DLCM reduces the number of model parameters and improves the accuracy of wildfire smoke detection. Then, the DySample upsampling operator is used in the upsampling part to make the image generation more accurate with very few parameters. Finally, in the course of the training process, the loss function is improved. EMASlideLoss is used to improve detection ability for small targets, and the Shape-IoU loss function is used to optimize the shape of wildfires and smoke. Experiments are conducted on wildfire and smoke datasets, and the final mAP50 is 86.6%, which is 1.5% higher than YOLOv8, and the number of parameters is decreased by 29.7%. The experimental findings demonstrate that LDENet is capable of effectively detecting wildfire smoke and ensuring the safety of transmission line corridors. Full article

To address challenges like manual processes, complicated detection methods, high false alarm rates, and frequent errors in identifying defects on steel surfaces, this research presents an innovative detection system, YOLOv10n-SFDC. The study focuses on the complex dependencies between parameters used for defect detection, particularly the interplay between feature extraction, fusion, and bounding box regression, which often leads to inefficiencies in traditional methods. YOLOv10n-SFDC incorporates advanced elements such as the DualConv module, SlimFusionCSP module, and Shape-IoU loss function, improving feature extraction, fusion, and bounding box regression to enhance accuracy. Testing on the NEU-DET dataset shows that YOLOv10n-SFDC achieves a mean average precision (mAP) of 85.5% at an Intersection over Union (IoU) threshold of 0.5, a 6.3 percentage point improvement over the baseline YOLOv10. The system uses only 2.67 million parameters, demonstrating efficiency. It excels in identifying complex defects like ’rolled in scale’ and ’inclusion’. Compared to SSD and Fast R-CNN, YOLOv10n-SFDC outperforms these models in accuracy while maintaining a lightweight architecture. This system excels in automated inspection for industrial environments, offering rapid, precise defect detection. YOLOv10n-SFDC emerges as a reliable solution for the continuous monitoring and quality assurance of steel surfaces, improving the reliability and efficiency of steel manufacturing processes. Full article

Accurate flow stress prediction is vital for optimizing the manufacturing of lightweight materials under high-temperature conditions. In this study, a boron nitride (BN)-reinforced AZ80 magnesium composite was subjected to hot compression tests at temperatures of 300–400 °C and strain rates ranging from 0.01 to 10 s⁻¹. A data-driven Support Vector Regression (SVR) model was developed to predict flow stress based on temperature, strain rate, and strain. Trained on experimental data, the SVR model demonstrated high predictive accuracy, as evidenced by a low mean squared error (MSE), a coefficient of determination (R²) close to unity, and a minimal average absolute relative error (AARE). Sensitivity analysis revealed that strain rate and temperature exerted the greatest influence on flow stress. By integrating machine learning with experimental observations, this framework enables efficient optimization of thermal deformation, supporting data-driven decision-making in forming processes. The results underscore the potential of combining advanced computational models with real-time experimental data to enhance manufacturing efficiency and improve process control in next-generation lightweight alloys. Full article

Existing surface defect detection methods for micro-motor commutators suffer from low detection accuracy, poor real-time performance, and high false detection and missed detection rates for small targets. To address these issues, this paper proposes a high-performance and robust commutator surface defect detection model (CLS-YOLO), using YOLOv11-n as the baseline model. First, a lightweight Cross-Scale Feature Fusion Module (CCFM) is introduced to integrate features from different scales, enhancing the model’s adaptability to scale variations and ability to detect small objects. This approach reduces model parameters and improves detection speed without compromising detection accuracy. Second, a Large Separable Kernel Attention (LSKA) module is incorporated into the detection head to strengthen feature understanding and capture, reducing interference from complex surface patterns on the commutator and significantly improving adaptability to various target types. Finally, to address issues related to the center point location, aspect ratio, angle, and sample imbalance in bounding boxes, SIoU Loss replaces the CIoU Loss in the original network, overcoming limitations of the original loss function and enhancing overall detection performance. Model performance was evaluated and compared on a commutator surface defect detection dataset, with additional experiments designed to verify the model’s effectiveness and feasibility. Experimental results show that, compared to YOLOv11-n, the CLS-YOLO model achieves a 2.08% improvement in [email protected]. This demonstrates that CLS-YOLO can accurately detect large defect targets while maintaining accuracy for tiny defects. Additionally, CLS-YOLO outperforms most YOLO-series models, including YOLOv8-n and YOLOv10-n. The model’s parameter count is only 1.860 million, lower than YOLOv11-n, with a detection speed increase of 8.34%, making it suitable for deployment on resource-limited terminal devices in complex industrial scenarios. Full article

The refinement of effective data generation methods has led to a growing interest in using artificial neural networks (ANNs) to solve modeling problems related to mechanical structures. This study investigates the modeling of composite sandwich structures, i.e., structures made up of two laminated composite face sheets sandwiching a lightweight honeycomb core. An ANN was utilized to predict structural deflection and face sheet stress with low computational cost. Initially, a three-point load mode was used to determine the flexural behavior of the composite sandwich structure before subsequently analyzing the sandwich structure using the Monte Carlo sampling tool. Various combinations of face sheet materials, face sheet layer numbers, core types, core thicknesses and load magnitudes were considered as design variables in data generation. The generated data were used to train a neural network. Subsequently, the predictions of the trained ANN were compared with the outcomes of a finite element model (FEM), and the comparison was extended to real structures by conducting experimental tests. A woven carbon-fiber-reinforced polymer (WCFRP) with a Nomex honeycomb core was tested to validate the ANN predictions. The predictions from the elaborated ANN model closely matched the FEM and experimental results. Therefore, this method offers a low-computational-cost technique for designing and optimizing sandwich structures in various engineering applications. Full article

As an emerging edge device aimed at consumers, Unmanned Aerial Vehicles (UAVs) have attracted significant attention in the consumer electronics market, particularly for intelligent imaging applications. However, aerial image detection tasks face two major challenges: first, there are numerous small and overlapping objects that are difficult to identify from an aerial perspective, and second, if the detection frame rate is not high enough, missed detections may occur when the UAV is moving quickly, which can negatively impact the user experience by reducing detection accuracy, increasing the likelihood of collision-avoidance failures, and potentially causing unsafe flight behavior. To address these challenges, this paper proposes a novel YOLO (you only look once) framework, named YOLOFLY, which includes a C4f feature extraction module and a DWcDetect head to make the model lightweight, as well as an MPSA attention mechanism and an ACIoU loss function, aimed at improving detection accuracy and performance for consumer-grade UAVs. Extensive experiments on the public VisDrone2019 dataset demonstrate that YOLOFLY outperforms the latest state-of-the-art model, YOLOv11n, by 3.2% in mAP_50-95, reduces detection time by 27.2 ms, decreases the number of parameters by 0.6 M, and cuts floating-point operations by 1.8 B. Finally, testing YOLOFLY in real-world environments also yielded the best results, including a 3.75% reduction in missed detections at high speeds. These findings validate the superiority and effectiveness of YOLOFLY. Full article