Search Results (2,712)

Breast cancer is one of the most prevalent malignant tumors among women worldwide, underscoring the urgent need for early and accurate diagnosis to reduce mortality. To address this, A Multi-Feature Fusion Classification Network (MFF-ClassificationNet) is proposed for breast histopathological image classification. The network adopts a two-branch parallel architecture, where a convolutional neural network captures local details and a Transformer models global dependencies. Their features are deeply integrated through a Multi-Feature Fusion module, which incorporates a Convolutional Block Attention Module—Squeeze and Excitation (CBAM-SE) fusion block combining convolutional block attention, squeeze-and-excitation mechanisms, and a residual inverted multilayer perceptron to enhance fine-grained feature representation and category-specific lesion characterization. Experimental evaluations on the BreakHis dataset achieved accuracies of 98.30%, 97.62%, 98.81%, and 96.07% at magnifications of 40×, 100×, 200×, and 400×, respectively, while an accuracy of 97.50% was obtained on the BACH dataset. These results confirm that integrating local and global features significantly strengthens the model’s ability to capture multi-scale and context-aware information, leading to superior classification performance. Overall, MFF-ClassificationNet surpasses conventional single-path approaches and provides a robust, generalizable framework for advancing computer-aided diagnosis of breast cancer. Full article

Vehicle re-identification (Re-ID) is a critical task in the fields of intelligent transportation and urban surveillance. This task faces numerous challenges, such as significant changes in shooting angles, strong similarities in appearance between different vehicles of the same model, and difficulties in modeling fine-grained differences. To overcome the shortcomings of existing methods in local feature extraction and multi-scale fusion, this paper proposes an attention-guided dual-branch feature fusion network (DAFF-Net). The network uses ResNet50-ibn as its backbone and designs two complementary feature extraction branches. One branch fuses cross-layer attention between shallow and deep features, introducing a Temperature-Calibration Attention Fusion Module (TCAF) to improve the accuracy of cross-layer feature fusion effectively. The other branch enhances multi-scale attention for mid-layer features, constructing a Multi-Scale Gated Attention Module (MSGA) to extract local details and directional structural information. Finally, the discriminative ability of the enhanced features is improved by concatenating the two branch features and jointly optimizing the network using triplet loss, cross-entropy loss, and center loss. Experimental results on the VeRi-776 and VehicleID public datasets indicate that the proposed DAFF-Net outperforms existing mainstream methods in multiple key metrics. On the VeRi-776 dataset, mAP and CMC@1 increased to 82.2% and 97.5%, respectively. In the three test subsets of the VehicleID dataset, the CMC@1 metric achieved 90.7%, 84.6%, and 82.1%, respectively, demonstrating the effectiveness of the proposed network in vehicle re-identification tasks. Full article

Unmanned Aerial Vehicle (UAV)-based aerial object detection has been widely applied in various fields, including logistics, public security, disaster response, and smart agriculture. However, numerous small objects in UAV aerial images are often overwhelmed by large-scale complex backgrounds, making their appearance difficult to distinguish and thereby prone to being missed by detectors. To tackle these issues, we propose a novel Contextual-Semantic Interactive Perception Network (CSIPN) for small object detection in UAV aerial scenarios, which enhances detection performance through scene interaction modeling, dynamic context modeling, and dynamic feature fusion. The core components of the CSIPN include the Scene Interaction Modeling Module (SIMM), the Dynamic Context Modeling Module (DCMM), and the Semantic-Context Dynamic Fusion Module (SCDFM). Specifically, the SIMM introduces a lightweight self-attention mechanism to generate a global scene semantic embedding vector, which then interacts with shallow spatial descriptors to explicitly depict the latent relationships between small objects and complex background, thereby selectively activating key spatial responses. The DCMM employs two dynamically adjustable receptive-field branches to adaptively model contextual cues and effectively supplement the contextual information required for detecting various small objects. The SCDFM utilizes a dual-weighting strategy to dynamically fuse deep semantic information with shallow contextual details, highlighting features relevant to small object detection while suppressing irrelevant background. Our method achieves mAPs of 37.2%, 93.4%, 50.8%, and 48.3% on the TinyPerson dataset, the WAID dataset, the VisDrone-DET dataset, and our self-built WildDrone dataset, respectively, while using only 25.3M parameters, surpassing existing state-of-the-art detectors and demonstrating its superiority and robustness. Full article

Rice leaf diseases are critical factors affecting rice yield and quality, and their effective detection is crucial for ensuring stable production. However, existing detection models exhibit limitations in capturing irregular and fine-grained lesion features and are susceptible to interference from complex backgrounds. To address these challenges, this study proposes an Efficient Attention-Guided Network (EAG-YOLOv11n) for rice leaf disease detection. Specifically, an EMA-C3K2 module is proposed to enhance the network’s feature extraction capability. It integrates Efficient Multi-Scale Attention (EMA) into the shallow C3K2 layers, enabling the network to extract richer low-level feature representations. In addition, a Global Local Complementary Attention module (GLC-PSA) is proposed, which integrates a Local Importance Attention (LIA) branch to enhance the local feature representation of the original C2PSA. This design strengthens the perception of lesion regions while effectively suppressing background interference. Furthermore, an Adaptive Threshold Focal Loss (ATFL) is employed to guide the optimization of model parameters during training, alleviating sample imbalance and adaptively emphasizing the learning of challenging samples. Experimental results demonstrate that EAG-YOLOv11n achieves a mean Average Precision (mAP) of 87.3%, representing a 2.7% improvement over the baseline. Furthermore, compared with existing mainstream detection methods, including RT-DETR-L, YOLOv7tiny, YOLOv8n, YOLOv9tiny, YOLOv10n, and YOLOv12n, EAG-YOLOv11n improves mAP by 4.2%, 4.3%, 1.9%, 1.1%, 8.7%, and 2.9%, respectively. Overall, these results highlight its superior effectiveness in rice leaf disease detection, providing reliable technical support for stable yield and sustainable agricultural development. Full article

In strong clutter and maneuvering scenarios, radar track initiation faces the dual challenges of a low initiation rate and high false alarm rate. Although the existing deep learning methods show promise, the commonly adopted “feature flattening” input strategy destroys the intrinsic temporal structure and feature relationships of track data, limiting its discriminative performance. To address this issue, this paper proposes a novel radar track initiation method based on Dual-Attention Temporal Convolutional Network (DA-TCN), reformulating track initiation as a binary classification task for very short multi-channel time series that preserve complete temporal structure. The DA-TCN model employs the TCN as its backbone network to extract local dynamic features and innovatively constructs a dual-attention architecture: a channel attention branch dynamically calibrates the importance of each kinematic feature, while a temporal attention branch integrates Bi-GRU and self-attention mechanisms to capture the dependencies at critical time steps. Ultimately, a learnable gated fusion mechanism adaptively weights the dual-branch information for optimal characterization of track characteristics. Experimental results on maneuvering target datasets demonstrate that the proposed method significantly outperforms multiple baseline models across varying clutter densities: Under the highest clutter density, DA-TCN achieves 95.12% true track initiation rate (+1.6% over best baseline) with 9.65% false alarm rate (3.63% reduction), validating its effectiveness for high-precision and highly robust track initiation in complex environments. Full article

Accurate delineation of glacier extent is crucial for monitoring glacier degradation in the context of global warming. Satellite remote sensing with high spatial and temporal resolution offers an effective approach for large-scale glacier mapping. However, persistent cloud cover limits its application on the Tibetan Plateau, leading to substantial omissions in glacier identification. Therefore, this study proposed a novel sub-cloudy glacier extraction model (SCGEM) designed to extract glacier boundaries from cloud-affected satellite images. First, the cloud-insensitive characteristics of topo-graphic (Topo.), synthetic aperture radar (SAR), and temporal (Tempo.) features were investigated for extracting glaciers under cloud conditions. Then, a transformer-based generative adversarial network (GAN) was proposed, which incorporates an image reconstruction and an adversarial branch to improve glacier extraction accuracy under cloud cover. Experimental results demonstrated that the proposed SCGEM achieved significant improvements with an IoU of 0.7700 and an F1 score of 0.8700. The Topo., SAR, and Tempo. features all contributed to glacier extraction in cloudy areas, with the Tempo. features contributing the most. Ablation studies further confirmed that both the adversarial training mechanism and the multi-task architecture contributed notably to improving the extraction accuracy. The proposed architecture serves both to data clean and enhance the extraction of glacier texture features. Full article

Insulator defect detection in power inspection tasks faces significant challenges due to the large variations in defect sizes and complex backgrounds, which hinder the accurate identification of both small and large defects. To overcome these issues, we propose a novel dual-branch YOLO-based algorithm (DB-YOLO), built upon the YOLOv11 architecture. The model introduces two dedicated branches, each tailored for detecting large and small defects, respectively, thereby enhancing robustness and precision across multiple scales. To further strengthen global feature representation, the Mamba mechanism is integrated, improving the detection of large defects in cluttered scenes. An adaptive weighted CIoU loss function, designed based on defect size, is employed to refine localization during training. Additionally, ShuffleNetV2 is embedded as a lightweight backbone to boost inference speed without compromising accuracy. We evaluate DB-YOLO on the following three datasets: the open source CPLID, a self-built insulator defect dataset, and GC-10. Experimental results demonstrate that DB-YOLO achieves superior performance in both accuracy and real-time efficiency compared to existing state-of-the-art methods. These findings suggest that the proposed approach offers strong potential for practical deployment in real-world power inspection applications. Full article

Event cameras provide microsecond temporal resolution, high dynamic range, and low latency by asynchronously capturing per-pixel luminance changes, thereby introducing a novel sensing paradigm. These advantages render them well-suited for high-speed applications such as autonomous vehicles and dynamic environments. Nevertheless, the sparsity of event data and the absence of dense annotations are significant obstacles to supervised learning for motion segmentation from event streams. Domain adaptation is also challenging due to the considerable domain shift in intensity images. To address these challenges, we propose a two-phase cross-modality adaptation framework that translates motion segmentation knowledge from labeled RGB-flow data to unlabeled event streams. A dual-branch encoder extracts modality-specific motion and appearance features from RGB and optical flow in the source domain. Using reconstruction networks, event voxel grids are converted into pseudo-image and pseudo-flow modalities in the target domain. These modalities are subsequently re-encoded using frozen RGB-trained encoders. Multi-level consistency losses are implemented on features, predictions, and outputs to enforce domain alignment. Our design enables the model to acquire domain-invariant, semantically rich features through the use of shallow architectures, thereby reducing training costs and facilitating real-time inference with a lightweight prediction path. The proposed architecture, alongside the utilized hybrid loss function, effectively bridges the domain and modality gap. We evaluate our method on two challenging benchmarks: EVIMO2, which incorporates real-world dynamics, high-speed motion, illumination variation, and multiple independently moving objects; and MOD++, which features complex object dynamics, collisions, and dense 1kHz supervision in synthetic scenes. The proposed UDA framework achieves 83.1% and 79.4% accuracy on EVIMO2 and MOD++, respectively, outperforming existing state-of-the-art approaches, such as EV-Transfer and SHOT, by up to 3.6%. Additionally, it is lighter and faster and also delivers enhanced mIoU and F1 Score. Full article

Accurate and reliable detection of landslides plays a crucial role in disaster prevention and mitigation efforts. However, due to unfavorable environmental conditions, uneven surface structures, and other disturbances similar to those of landslides, traditional methods often fail to achieve the desired results. To address these challenges, this study introduces a novel multi-scale cross-layer feature interaction network, specifically designed for landslide segmentation in remote sensing images. In the MCFI-Net framework, we adopt the encoder–decoder as the foundational architecture, and integrate cross-layer feature information to capture fine-grained local textures and broader contextual patterns. Then, we introduce the receptive field block (RFB) into the skip connections to effectively aggregate multi-scale contextual information. Additionally, we design the multi-branch dynamic convolution block (MDCB), which possesses both dynamic perception ability and multi-scale feature representation capability. The comprehensive evaluation conducted on both the Landslide4Sense and Bijie datasets demonstrates the superior performance of MCFI-Net in landslide segmentation tasks. Specifically, on the Landslide4Sense dataset, MCFI-Net achieved a Dice score of 0.7254, a Matthews correlation coefficient (Mcc) of 0.7138, and a Jaccard score of 0.5699. Similarly, on the Bijie dataset, MCFI-Net maintained high accuracy with a Dice score of 0.8201, an Mcc of 0.8004, and a Jaccard score of 0.6951. Furthermore, when evaluated on the optical remote sensing dataset EORSSD, MCFI-Net obtained a Dice score of 0.7770, an Mcc of 0.7732, and a Jaccard score of 0.6571. Finally, ablation experiments carried out on the Landslide4Sense dataset further validated the effectiveness of each proposed module. These results affirm MCFI-Net’s capability in accurately identifying landslide regions from complex remote sensing imagery, and it provides great potential for the analysis of geological disasters in the real world. Full article

The inversion of shallow underground vibration fields primarily relies on signals collected by numerous sensors deployed on the surface. However, the accuracy of inversion is affected by the spatial distribution of these sensors. Therefore, under limited measurement points, signal reconstruction at unknown locations remains a critical challenge. To address this problem, we developed an SC-Net-based self-supervised interpolation method for missing wavefield data in shallow subsurface applications. This study utilizes incomplete seismic data acquired in real-world scenarios to train a neural network for seismic data interpolation, thereby expanding the sampled signals required for inversion. Since available seismic data samples are often scarce in practice, we adopt a hybrid training strategy combining simulated and real data. Specifically, a large number of numerically simulated samples are jointly trained with a limited set of real-world measurements. Furthermore, to enhance the robustness of network outputs, we integrate the Mean Teacher model framework and propose a self-supervised learning approach for missing data. Additionally, to enable the network to effectively capture long-range dependencies in both frequency and spatial domains of seismic data, we introduce a dual-branch feature fusion network that jointly models channel-wise and spatial relationships. Finally, in our actual field explosion experiments conducted at the test site, we demonstrated improved accuracy of our method through comparative analysis with several typical interpolation neural networks. Three ablation studies are also designed to demonstrate the effectiveness of the proposed approach. Full article

To mitigate the reduced accuracy of direction-of-arrival (DOA) estimation in scenarios with low signal-to-noise ratios (SNR) and multiple interfering sources, this paper proposes an Auxiliary Classifier Generative Adversarial Network (ACGAN) architecture that integrates a Squeeze-and-Excitation (SE) attention mechanism and a Multi-scale Dilated Feature Aggregation (MDFA) module. In this neural network, a vector hydrophone array is employed as the receiving unit, capable of simultaneously sensing particle velocity signals in three directions $\left(v_x,v_y,v_z\right)$ and acoustic pressure $p$, thereby providing high directional sensitivity and maintaining robust classification performance under low-SNR conditions. The MDFA module extracts features from multiple receptive fields, effectively capturing cross-scale patterns and enhancing the representation of weak targets in beamforming maps. This helps mitigate estimation bias caused by mutual interference among multiple targets in low-SNR environments. Furthermore, an auxiliary classification branch is incorporated into the discriminator to jointly optimize generation and classification tasks, enabling the model to more effectively identify and separate multiple types of labeled sources. Experimental results indicate that the proposed network is effective and shows improved performance across diverse scenarios. Full article

We propose a novel meta-learning enhanced 3D CNN-LSTM framework for durability prediction. The framework integrates 3D microstructural data from micro-CT scanning with environmental time-series data through a dual-branch architecture: a 3D CNN branch extracts spatial degradation patterns from volumetric data, while an LSTM network processes temporal environmental factors. To address data scarcity, we incorporate a prototypical network-based meta-learning module that learns class prototypes from limited support samples and generalizes predictions to new corrosion scenarios through distance-based probability estimation. Additionally, we develop a dynamic feature fusion mechanism that adaptively combines spatial, environmental, and mechanical features using trainable attention coefficients, enabling context-aware representation learning. Finally, an interface damage visualization component identifies critical degradation zones and propagation trajectories, providing interpretable engineering insights. Experimental validation on laboratory specimens demonstrates superior accuracy (74.6% in 1-shot scenarios) compared to conventional methods, particularly in aggressive corrosion environments where data scarcity typically hinders reliable prediction. The visualization system generates interpretable 3D damage maps with an average Intersection-over-Union of 0.78 compared to ground truth segmentations. This work establishes a unified computational framework bridging microstructure analysis with macroscopic durability assessment, offering practical value for infrastructure maintenance decision-making under uncertainty. The modular design facilitates extension to diverse interface types and environmental conditions. Full article

In the semantic segmentation tasks of polarimetric SAR images, deep learning has become an important end-to-end method that uses convolutional neural networks (CNNs) and other advanced network architectures to extract features and classify the target region pixel by pixel. However, applying original networks used to optical images for PolSAR image segmentation directly will result in the loss of rich phase information in PolSAR data, which leads to unsatisfactory classification results. In order to make full use of polarization information, the complex-valued BiSeNetV2 with a bilateral-segmentation structure is studied and expanded in this work. Then, considering further improving the ability to extract semantic features in the complex domain and alleviating the imbalance of polarization channel response, the complex-valued BiSeNetV2 with a lightweight attention module (LAM-CV-BiSeNetV2) is proposed for the semantic segmentation of PolSAR images. LAM-CV-BiSeNetV2 supports complex-valued operations, and a lightweight attention module (LAM) is designed and introduced at the end of the Semantic Branch to enhance the extraction of detailed features. Compared with the original BiSeNetV2, the LAM-CV-BiSeNetV2 can not only more fully extract the phase information from polarimetric SAR data, but also has stronger semantic feature extraction capabilities. The experimental results on the Flevoland and San Francisco datasets demonstrate that the proposed LAM has better and more stable performance than other commonly used attention modules, and the proposed network can always obtain better classification results than BiSeNetV2 and other known real-valued networks. Full article

This study develops a discrete element model incorporating the water–ice phase transition volume effect to simulate frost damage in saturated granite. The model investigates the damage evolution and mechanical degradation under freeze–thaw cycles. The results show that during freeze–thaw cycles, the model’s temperature field exhibits non-uniform distribution characteristics and geometric dependency, with lower maximum temperature differences in Brazilian disk models versus uniaxial compression specimens. Frost heave damage progresses through three distinct stages: localized bond fractures (1~5 cycles); accelerated crack interconnection and branching (15~20 cycles); and fully interconnected damage zones (25~30 cycles). As the number of freeze–thaw cycles increases, the crack network significantly influences the mechanical behavior of the model under load. The failure mode of the loaded model undergoes a transformation from brittle penetration to ductile fragmentation. Freeze–thaw cycles cause more significant degradation in the tensile strength of granite compared to compressive strength. After 30 freeze–thaw cycles, the uniaxial compressive strength and Brazilian tensile strength decrease by 47.5% and 93.8%, respectively. These findings provide theoretical support for assessing frost heave damage in geotechnical engineering in cold regions. Full article

Person re-identification (re-ID) aims to retrieve images of a given individual from different camera views. Obstacles obstructing parts of a pedestrian’s body often result in incomplete identity information, impairing recognition performance. To address the occlusion problem, a method called Multi-Branch Interaction Network (MBIN) is proposed, which exploits the information interaction between different branches to effectively characterize occluded pedestrians for person re-ID. The method consists primarily of a hard branch, a soft branch, and a view branch. The hard branch enhances feature robustness via a unified horizontal partitioning strategy. The soft branch improves the high-level feature representation via multi-head attention. The view branch fuses multi-view feature maps to form a comprehensive representation via a dual-classifier fusion mechanism. Moreover, a mutual knowledge distillation strategy is employed to promote knowledge exchange among the three branches. Extensive experiments are conducted on widely used person re-ID datasets to validate the effectiveness of our method. Full article