Search Results (802)

Building extraction from high-resolution remote sensing imagery is critical for urban planning and disaster management, yet remains challenging due to significant intra-class variability in architectural styles and multi-scale distribution patterns of buildings. To address these limitations, we propose the Multi-Scale Guided Context-Aware Network (MSGCANet), a Transformer-based multi-scale guided context-aware network. Our framework integrates a Contextual Exploration Module (CEM) that synergizes asymmetric and progressive dilated convolutions to hierarchically expand receptive fields, enhancing discriminability for dense building features. We further design a Window-Guided Multi-Scale Attention Mechanism (WGMSAM) to dynamically establish cross-scale spatial dependencies through adaptive window partitioning, enabling precise fusion of local geometric details and global contextual semantics. Additionally, a cross-level Transformer decoder leverages deformable convolutions for spatially adaptive feature alignment and joint channel-spatial modeling. Experimental results show that MSGCANet achieves IoU values of 75.47%, 91.53%, and 83.10%, and F1-scores of 86.03%, 95.59%, and 90.78% on the Massachusetts, WHU, and Inria datasets, respectively, demonstrating robust performance across these datasets. Full article

In semantic segmentation tasks, large kernels and Atrous convolution have been utilized to increase the receptive field, enabling models to achieve competitive performance with fewer parameters. However, due to the fixed size of kernel functions, networks incorporating large convolutional kernels are limited in adaptively capturing multi-scale features and fail to effectively leverage global contextual information. To address this issue, we combine Atrous convolution with large kernel convolution, using different dilation rates to compensate for the single-scale receptive field limitation of large kernels. Simultaneously, we employ a dynamic selection mechanism to adaptively highlight the most important spatial features based on global information. Additionally, to enhance the model’s ability to fit the true label distribution, we propose a Multi-Scale Contextual Noise Transfer Matrix (NTM), which uses high-order consistency information from neighborhood representations to estimate NTM and correct supervision signals, thereby improving the model’s generalization capability. Extensive experiments conducted on Cityscapes, ADE20K, and COCO-Stuff-10K demonstrate that this approach achieves a new state-of-the-art balance between speed and accuracy. Specifically, LKNTNet achieves 80.05% mIoU on Cityscapes with an inference speed of 80.7 FPS and 42.7% mIoU on ADE20K with an inference speed of 143.6 FPS. Full article

This paper proposes Feature fusion and Dilated causal convolution model for Bearing Remaining useful life Prediction (FDBRP), an integrated framework for accurate Remaining Useful Life (RUL) prediction of rolling bearings that combines three key innovations: (1) a data augmentation module employing sliding-window processing and two-dimensional feature concatenation with label normalization to enhance signal representation and improve model generalizability, (2) a feature fusion module incorporating an enhanced graph convolutional network for spatial modeling, an improved multi-scale temporal convolution for dynamic pattern extraction, and an efficient multi-scale attention mechanism to optimize spatiotemporal feature consistency, and (3) an optimized dilated convolution module utilizing interval sampling to expand the receptive field, and combines the residual connection structure to realize the regularization of the neural network and enhance the ability of the model to capture long-range dependencies. Experimental validation showcases the effectiveness of proposed approach, achieving a high average score of 0.756564 and demonstrating a lower average error of 10.903656 in RUL prediction for test bearings compared to state-of-the-art benchmarks. This highlights the superior RUL prediction capability of the proposed methodology. Full article

To address the challenges of low diagnostic accuracy arising from the non-stationary and nonlinear time-varying characteristics of acoustic signals in rolling bearing fault diagnosis, as well as their susceptibility to noise interference, this paper proposes a fault diagnosis method based on a Weighted Average Algorithm–Feature Mode Decomposition (WAA-FMD) and a Local–Global Adaptive Multi-scale Attention Mechanism (LGAF)–Swin Transformer. First, the WAA is utilized to optimize the key parameters of FMD, thereby enhancing its signal decomposition performance while minimizing noise interference. Next, a bilateral expansion strategy is implemented to extend both the time window and frequency band of the signal, which improves the temporal locality and frequency globality of the time–frequency diagram, significantly enhancing the ability to capture signal features. Ultimately, the introduction of depthwise separable convolution optimizes the receptive field and improves the computational efficiency of shallow networks. When combined with the Swin Transformer, which incorporates LGAF and adaptive feature selection modules, the model further enhances its perceptual capabilities and feature extraction accuracy through dynamic kernel adjustment and deep feature aggregation strategies. The experimental results indicate that the signal denoising performance of WAA-FMD significantly outperforms traditional denoising techniques. In the KAIST dataset (NSK 6205: inner raceway fault and outer raceway fault) and the experimental dataset (FAG 30205: inner raceway fault, outer raceway fault, and rolling element fault), the accuracies of the proposed model reach 100% and 98.62%, respectively, both exceeding that of other deep learning models. In summary, the proposed method demonstrates substantial advantages in noise reduction performance and fault diagnosis accuracy, providing valuable theoretical insights for practical applications. Full article

To precisely delineate the contours of insulators in complex transmission line images obtained from Unmanned Aerial Vehicle (UAV) inspections and thereby facilitate subsequent defect analysis, this study proposes an instance segmentation framework predicated upon an enhanced SOLOv2 model. The proposed framework integrates a preprocessed edge channel, generated through the Non-Subsampled Contourlet Transform (NSCT), which augments the model’s capability to accurately capture the edges of insulators. Moreover, the input image resolution to the network is heightened to 1200 × 1600, permitting more detailed extraction of edges. Rather than the original ResNet + FPN architecture, the improved HRNet is utilized as the backbone to effectively harness multi-scale feature information, thereby enhancing the model’s overall efficacy. In response to the increased input size, there is a reduction in the network’s channel count, concurrent with an increase in the number of layers, ensuring an adequate receptive field without substantially escalating network parameters. Additionally, a Convolutional Block Attention Module (CBAM) is incorporated to refine mask quality and augment object detection precision. Furthermore, to bolster the model’s robustness and minimize annotation demands, a virtual dataset is crafted utilizing the fourth-generation Unreal Engine (UE4). Empirical results reveal that the proposed framework exhibits superior performance, with AP0.50 (90.21%), AP0.75 (83.34%), and AP[0.50:0.95] (67.26%) on a test set consisting of images supplied by the power grid. This framework surpasses existing methodologies and contributes significantly to the advancement of intelligent transmission line inspection. Full article

Accurate monitoring of snow and ice cover is essential for climate research and disaster management, but conventional remote sensing methods often struggle in complex terrain and fog-contaminated conditions. To address the challenges of high-resolution UAV-based snow and ice segmentation—including visual similarity, fragmented spatial distributions, and terrain shadow interference—we introduce IceSnow-Net, a U-Net-based architecture enhanced with three key components: (1) a ResNet50 backbone with atrous convolutions to expand the receptive field, (2) an Atrous Spatial Pyramid Pooling (ASPP) module for multi-scale context aggregation, and (3) an auxiliary path loss for deep supervision to enhance boundary delineation and training stability. The model was trained and validated on UAV-captured orthoimagery from Ganzi Prefecture, Sichuan, China. The experimental results demonstrate that IceSnow-Net achieved excellent performance compared to other models, attaining a mean Intersection over Union (mIoU) of 98.74%, while delivering 27% higher computational efficiency than U-Mamba. Ablation studies further validated the individual contributions of each module. Overall, IceSnow-Net provides an effective and accurate solution for cryosphere monitoring in topographically complex environments using UAV imagery. Full article

Deep learning has excelled in image classification largely due to large, professionally labeled datasets. However, in the field of medical images data annotation often relies on experienced experts, especially in tasks such as white blood cell classification where the staining methods for different cells vary greatly and the number of samples in certain categories is relatively small. To evaluate leukocyte classification performance with limited labeled samples, a few-shot learning method based on Feature Reconstruction Network with Improved EfficientNetV2 (FRNE) is proposed. Firstly, this paper presents a feature extractor based on the improved EfficientNetv2 architecture. To enhance the receptive field and extract multi-scale features effectively, the network incorporates an ASPP module with dilated convolutions at different dilation rates. This enhancement improves the model’s spatial reconstruction capability during feature extraction. Subsequently, the support set and query set are processed by the feature extractor to obtain the respective feature maps. A feature reconstruction-based classification method is then applied. Specifically, ridge regression reconstructs the query feature map using features from the support set. By analyzing the reconstruction error, the model determines the likelihood of the query sample belonging to a particular class, without requiring additional modules or extensive parameter tuning. Evaluated on the LDWBC and Raabin datasets, the proposed method achieves accuracy improvements of 3.67% and 1.27%, respectively, compared to the method that demonstrated strong OA performance on both datasets among all compared approaches. Full article

Power equipment detection is a critical component in power transmission line inspection. However, existing power equipment detection algorithms often face problems such as large model sizes and high computational complexity. This paper proposes a lightweight power equipment detection algorithm based on large receptive field and attention guidance. First, we propose a lightweight large receptive field feature extraction module, CRepLK, which reparameterizes multiple branches into large kernel convolution to improve the multi-scale detection capability of the model; secondly, we propose a lightweight ELA-guided Dynamic Sampling Fusion (LEDSF) Neck, which alleviates the feature misalignment problem inherent in conventional neck networks to a certain extent; finally, we propose a lightweight Partial Asymmetric Detection Head (PADH), which utilizes the redundancy of feature maps to achieve the significant light weight of the detection head. Experimental results show that on the Insplad power equipment dataset, the number of parameters, computational cost (GFLOPs) and the size of the model weight are reduced by 46.8%, 44.1% and 46.4%, respectively, compared with the Baseline model, while the mAP is improved by 1%. Comparative experiments on three power equipment datasets show that our model achieves a compelling balance between efficiency and detection performance in power equipment detection scenarios. Full article

Aero-engines, as complex systems integrating numerous rotating components and accessory equipment, operate under harsh and demanding conditions. Prolonged use often leads to frequent mechanical wear and surface defects on accessory parts, which significantly compromise the engine’s normal and stable performance. Therefore, accurately and rigorously identifying failure modes is of critical importance. In this study, failure modes are categorized into notches, scuffs, and scratches based on original bearing structure images. The YOLOv8 architecture is adopted as the base framework, and a Dilated Reparameterization Block (DRB) is introduced to enhance the Dilation-Wise Residual (DWR) module. This structure uses a large convolutional kernel to capture fragmented and sparse features in wear images, ensuring a wide receptive field. The concept of structural reparameterization is incorporated into DWR to improve its ability to capture detailed target information. Additionally, the standard convolutional layer in the head of the improved DWR-DRB structure is replaced by Spatial-Depth Convolution (SPD-Conv) to reduce the loss of wear morphology and enhance the accuracy of fault feature extraction. Finally, a fusion structure combining Focaler and MPDIoU is integrated into the loss function to leverage their strengths in handling imbalanced classification and bounding box geometric regression. The proposed method achieves effective recognition and diagnosis of mechanical wear fault patterns. The experimental results demonstrate that, compared to the baseline YOLOv8, the proposed method improves the mAP50 for fault diagnosis and recognition from 85.4% to 91%. Full article

Downhole torsional vibrations, especially high-frequency torsional oscillations (HFTOs) and stick–slip phenomena, pose a serious threat to drilling operations, often resulting in tool damage, prolonged non-productive time, and significant cost increases. Traditional monitoring methods cannot promptly capture complex vibration patterns, so there is an urgent need for advanced early warning systems. This study proposes the DS-DW-TimesNet model, which improves the TimesNet framework by incorporating downsampling technology for efficient data compression, dilated convolution that can expand the temporal receptive field, and a learnable weight normalization method that can stabilize the training process, thereby enhancing the capabilities of feature extraction and long-sequence modeling. Verified using field data from the Fuman Oilfield, the results show that in terms of the mean absolute error (MAE) for 210 s predictions, this model is 77.2% and 21.8% lower than LSTM and Informer, respectively, and the inference speed is increased by 78.5% (reaching 48 milliseconds). It can provide reliable 210 s early warning windows for high-frequency torsional oscillations and 150 s early warning windows for stick–slip, exceeding industry standards and helping to improve the safety and efficiency of drilling operations. Full article

Automatic skin lesion segmentation is essential for early melanoma diagnosis, yet the scarcity and limited diversity of annotated training data hinder progress. We introduce a two-stage framework that first employs a denoising diffusion probabilistic model (DDPM) enhanced with dilated convolutions and self-attention to synthesize unseen, high-fidelity dermoscopic images. In the second stage, segmentation models—including a dilated U-Net variant that leverages dilated convolutions to enlarge the receptive field—are trained on the augmented dataset. Experimental results demonstrate that this approach not only enhances segmentation accuracy across various architectures with an increase in DICE of more than 0.4, but also enables compact and computationally efficient segmentation models to achieve performance comparable to or even better than that of models with 10 times the parameters. Moreover, our diffusion-based data augmentation strategy consistently improves segmentation performance across multiple architectures, validating its effectiveness for developing accurate and deployable clinical tools. Full article

(1) Objective: This study is aiming at the key problems, such as insufficient detection accuracy of small targets and complex background interference in remote-sensing image target detection; (2) Methods: by optimizing the YOLOv10x model architecture, the YOLO-KRM model is proposed. Firstly, a new backbone network structure is constructed. By replacing the C2f of the third layer of the backbone network with the Kolmogorov–Arnold network, the approximation ability of the model to complete complex nonlinear functions in high-dimensional space is improved. Then, the C2f of the fifth layer of the backbone network is replaced by the receptive field attention convolution, which enhances the model’s ability to capture the global context information of the features. In addition, the C2f and C2fCIB structures in the upsampling operation in the neck network are replaced by the hybrid local channel attention mechanism module, which significantly improves the feature representation ability of the model. Results: In order to validate the effectiveness of the YOLO-KRM model, detailed experiments were conducted on two remote-sensing datasets, RSOD and NWPU VHR-10. The experimental results show that, compared with the original model YOLOv10x, the mAP@50 of the YOLO-KRM model on the two datasets is increased by 1.77% and 2.75%, respectively, and the mAP @ 50:95 index is increased by 3.82% and 5.23%, respectively; (3) Results: by improving the model, the accuracy of target detection in remote-sensing images is successfully enhanced. The experimental results verify the effectiveness of the model in dealing with complex backgrounds and small targets, especially in high-resolution remote-sensing images. Full article

Motion blur is a complex phenomenon caused by the relative movement between an observed object and an imaging sensor during the exposure time, resulting in degradation in the image quality. Deep-learning-based methods, particularly convolutional neural networks (CNNs), have shown promise in motion deblurring. However, the small kernel sizes of CNNs limit their ability to achieve optimal performance. Moreover, supervised deep-learning-based deblurring methods often exhibit overfitting in their training datasets. Models trained on widely used synthetic blur datasets frequently fail to generalize in other blur domains in real-world scenarios and often produce undesired artifacts. To address these challenges, we propose the Spatial Feature Selection Network (SFSNet), which incorporates a Regional Feature Extractor (RFE) module to expand the receptive field and effectively select critical spatial features in order to improve the deblurring performance. In addition, we present the BlurMix dataset, which includes diverse blur types, as well as a meta-tuning strategy for effective blur domain adaptation. Our method enables the network to rapidly adapt to novel blur distributions with minimal additional training, and thereby improve generalization. The experimental results show that the meta-tuning variant of the SFSNet eliminates unwanted artifacts and significantly improves the deblurring performance across various blur domains. Full article

Anomaly detection is inherently challenging, as anomalies typically emerge only at test time. While reconstruction-based methods are popular, their reliance on CNN backbones with local receptive fields limits discrimination and precise localization. We propose FFC-AD, a reconstruction framework using Fourier Feature Convolutions (FFCs) to capture global information early, and we introduce Hidden Space Anomaly Simulation (HSAS), a latent-space regularization strategy that mitigates overgeneralization. Experiments on MVTec AD and VisA demonstrate that FFC-AD significantly outperforms state-of-the-art methods in both detection and segmentation accuracy. Full article

Limited depth of field in modern optical imaging systems often results in partially focused images. Multi-focus image fusion (MFF) addresses this by synthesizing an all-in-focus image from multiple source images captured at different focal planes. While deep learning-based MFF methods have shown promising results, existing approaches face significant challenges. Convolutional Neural Networks (CNNs) often struggle to capture long-range dependencies effectively, while Transformer and Mamba-based architectures, despite their strengths, suffer from high computational costs and rigid input size constraints, frequently necessitating patch-wise fusion during inference—a compromise that undermines the realization of a true global receptive field. To overcome these limitations, we propose MLP-MFF, a novel lightweight, end-to-end MFF network built upon the Pyramid Fusion Multi-Layer Perceptron (PFMLP) architecture. MLP-MFF is specifically designed to handle flexible input scales, efficiently learn multi-scale feature representations, and capture critical long-range dependencies. Furthermore, we introduce a Dual-Path Adaptive Multi-scale Feature-Fusion Module based on Hybrid Attention (DAMFFM-HA), which adaptively integrates hybrid attention mechanisms and allocates weights to optimally fuse multi-scale features, thereby significantly enhancing fusion performance. Extensive experiments on public multi-focus image datasets demonstrate that our proposed MLP-MFF achieves competitive, and often superior, fusion quality compared to current state-of-the-art MFF methods, all while maintaining a lightweight and efficient architecture. Full article