Search Results (98)

Named entity recognition (NER) is one of the fundamental tasks in building knowledge graphs. For some domain-specific corpora, the text descriptions exhibit limited standardization, and some entity structures have entity nesting. The existing entity recognition methods have problems such as word matching noise interference and difficulty in distinguishing different entity labels for the same character in sequence label prediction. This paper proposes a span-based feature reuse stacked bidirectional long short term memory network (BiLSTM) nested named entity recognition (SFRSN) model, which transforms the entity recognition of sequence prediction into the problem of entity span suffix category classification. Firstly, character feature embedding is generated through bidirectional encoder representation of transformers (BERT). Secondly, a feature reuse stacked BiLSTM is proposed to obtain deep context features while alleviating the problem of deep network degradation. Thirdly, the span feature is obtained through the dilated convolution neural network (DCNN), and at the same time, a single-tail selection function is introduced to obtain the classification feature of the entity span suffix, with the aim of reducing the training parameters. Fourthly, a global feature gated attention mechanism is proposed, integrating span features and span suffix classification features to achieve span suffix classification. The experimental results on four Chinese-specific domain datasets demonstrate the effectiveness of our approach: SFRSN achieves micro-F1 scores of 83.34% on ontonotes, 73.27% on weibo, 96.90% on resume, and 86.77% on the supply chain management dataset. This represents a maximum improvement of 1.55%, 4.94%, 2.48%, and 3.47% over state-of-the-art baselines, respectively. The experimental results demonstrate the effectiveness of the model in addressing nested entities and entity label ambiguity issues. Full article

In real-world visual recognition tasks, long-tailed distributions pose a widespread challenge, with extreme class imbalance severely limiting the representational learning capability of deep models. In practice, due to this imbalance, deep models often exhibit poor generalization performance on tail classes. To address this issue, data augmentation through the synthesis of new tail-class samples has become an effective method. One popular approach is CutMix, which explicitly mixes images from tail and other classes, constructing labels based on the ratio of the regions cropped from both images. However, region-based labels completely ignore the inherent semantic information of the augmented samples. To overcome this problem, we propose a saliency-guided local semantic mixing (LSM) method, which uses differentiable block decoupling and semantic-aware local mixing techniques. This method integrates head-class backgrounds while preserving the key discriminative features of tail classes and dynamically assigns labels to effectively augment tail-class samples. This results in efficient balancing of long-tailed data distributions and significant improvements in classification performance. The experimental validation shows that this method demonstrates significant advantages across three long-tailed benchmark datasets, improving classification accuracy by 5.0%, 7.3%, and 6.1%, respectively. Notably, the LSM framework is highly compatible, seamlessly integrating with existing classification models and providing significant performance gains, validating its broad applicability. Full article

In recent years, tail-sitter unmanned aerial vehicles (UAVs), capable of vertical take-off and landing (VTOL) and long-range flight, have garnered extensive attention. However, the challenge of yaw control, particularly for large-scale UAVs, has become a significant obstacle. It is challenging to generate sufficient yaw moments by motor differential thrust without compromising control authority in other channels or increasing mechanical complexity. Therefore, this paper proposes the concept of blown yaw, which utilizes the high-velocity airflow over rudders, induced by the propellers slipstream, to enhance the yaw control torque actively. An over-actuated, hundred-kilogram-class, tail-sitter UAV is designed to validate the effectiveness of the proposed method. To address the control allocation problem introduced by the implementation of blown yaw, an optimization-based control allocation module is developed, capable of precisely mapping the required forces and torques to all actuators. The proposed method, combined with computational fluid dynamics (CFD) simulations, accounts for the propeller model and the significant differences in actuator effectiveness across various flight conditions. Simulation results demonstrate that the proposed blown-yaw method significantly enhances the yaw control performance, achieving an overall energy savings of approximately 8.0% and a 60% reduction in the mean-squared error. Furthermore, the method exhibits robust performance against variations in control parameters and external disturbances. Full article

Panoptic scene graph generation (PSG) aims to simultaneously segment both foreground objects and background regions while predicting object relations for fine-grained scene modeling. Despite significant progress in panoptic scene understanding, current PSG methods face challenging problems: relation prediction often only relies on visual representations and is hindered by imbalanced relation category distributions. Accordingly, we propose IntegraPSG, a single-stage framework that integrates large language model (LLM) guidance with multimodal feature fusion. IntegraPSG introduces a multimodal sparse relation prediction network that efficiently integrates visual, linguistic, and depth cues to identify subject–object pairs most likely to form relations, enhancing the screening of subject–object pairs and filtering dense candidates into sparse, effective pairs. To alleviate the long-tail distribution problem of relations, we design a language-guided multimodal relation decoder where LLM is utilized to generate language descriptions for relation triplets, which are cross-modally attended with vision pair features. This design enables more accurate relation predictions for sparse subject–object pairs and effectively improves discriminative capability for rare relations. Experimental results show that IntegraPSG achieves steady and strong performance on the PSG dataset, especially with the R@100, mR@100, and mean reaching 38.7%, 28.6%, and 30.0%, respectively, indicating strong overall results and supporting the validity of the proposed method. Full article

As a legacy of historical metal mining and the processing and smelting of metalliferous ores, metal pollution is a serious environmental problem in many areas around the globe. This review summarizes the history, technical development and environmental hazards of historic metal mining and metallurgical activities in the Harz Region (Germany), one of the oldest and most productive mining landscapes in Central Europe. The release of large amounts of metal-containing waste into rivers during historic ore processing and the ongoing leaching of metals from slag heaps, tailings dumps and contaminated soils and sediments are the main sources of metal pollution in the Harz Mountains and its foreland. This pollution extends along river systems with tributaries from the Harz Mountains and can even be detected in mudflats of the North Sea. In addition to fluvial discharges, atmospheric pollution by smelter smoke has led to long-term damage to soils and vegetation in the Harz Region. Currently, the ecological hazards caused by the legacy pollution from historical metal mining and metallurgy in the Harz Region are only partially known, particularly regarding the effects of changes in river ecosystems as a consequence of climate change. This review discusses the complexity and dynamics of human–environment interactions in the Harz Mountains and its surroundings, with a focus on lead (Pb) pollution. The paper also identifies future research directions with respect to metal contamination. Full article

With the rapid development of remote sensing technology, satellites can now capture multispectral (MS) and panchromatic (PAN) images simultaneously. MS images offer rich spectral details, while PAN images provide high spatial resolutions. Effectively leveraging their complementary strengths and addressing modality gaps are key challenges in improving the classification performance. From the perspective of deep learning, this paper proposes a novel dual-source remote sensing classification framework named the Diversity Extraction and Fusion Classifier (DEFC-Net). A central innovation of our method lies in introducing a modality-specific intra-class diversity modeling mechanism for the first time in dual-source classification. Specifically, the intra-class diversity identification and splitting (IDIS) module independently analyzes the intra-class variance within each modality to identify semantically broad classes, and it applies an optimized K-means method to split such classes into fine-grained sub-classes. In particular, due to the inherent representation differences between the MS and PAN modalities, the same class may be split differently in each modality, allowing modality-aware class refinement that better captures fine-grained discriminative features in dual perspectives. To handle the class imbalance introduced by both natural long-tailed distributions and class splitting, we design a long-tailed ensemble learning module (LELM) based on a multi-expert structure to reduce bias toward head classes. Furthermore, a dual-modal knowledge distillation (DKD) module is developed to align cross-modal feature spaces and reconcile the label inconsistency arising from modality-specific class splitting, thereby facilitating effective information fusion across modalities. Extensive experiments on datasets show that our method significantly improves the classification performance. The code was accessed on 11 April 2025. Full article

In structural health monitoring (SHM), ensuring data completeness is critical for enhancing the accuracy and reliability of structural condition assessments. SHM data are prone to random missing values due to signal interference or connectivity issues, making precise data imputation essential. A latent factorization of tensor (LFT)-based method has proven effective for such problems, with optimization typically achieved via stochastic gradient descent (SGD). However, SGD-based LFT models and other imputation methods exhibit significant sensitivity to learning rates and slow tail-end convergence. To address these limitations, this study proposes an RMSprop-incorporated latent factorization of tensor (RLFT) model, which integrates an adaptive learning rate mechanism to dynamically adjust step sizes based on gradient magnitudes. Experimental validation on a scaled bridge accelerometer dataset demonstrates that RLFT achieves faster convergence and higher imputation accuracy compared to state-of-the-art models including SGD-based LFT and the long short-term memory (LSTM) network, with improvements of at least 10% in both imputation accuracy and convergence rate, offering a more efficient and reliable solution for missing data handling in SHM. Full article

Class imbalance (CI) poses a significant challenge in machine learning, characterized by a substantial disparity in sample sizes between majority and minority classes, leading to a pronounced “long-tail effect” in statistical distributions and subsequent inference processes. This issue is particularly acute in high-resolution land cover classification within arid regions, where CI tends to bias classification outcomes towards majority classes, often at the expense of minority classes. Recent advancements in deep learning have opened new avenues for tackling the CI problem in this context, focusing on three key aspects: the semantic segmentation model, loss function design, and dataset composition. To address this issue, we propose the high-resolution U-shaped mamba network (HRUMamba), which integrates multiple innovations to enhance segmentation performance under imbalanced conditions. Specifically, HRUMamba adopts a pre-trained HRNet as the encoder for capturing fine-grained local features and incorporates a modified scaled visual state space (SVSS) block in the decoder to model long-range dependencies effectively. An adaptive awareness fusion (AAF) module is embedded within the skip connections to enhance target saliency. Additionally, we introduce a synthetic loss function that combines cross-entropy loss, Dice loss, and auxiliary loss to improve optimization stability. To quantitatively assess multi-class imbalance, we introduce the coefficient of variation (CV) as a novel evaluation metric. Experimental results on the ISPRS Vaihingen and Minqin datasets demonstrate the robustness and effectiveness of HRUMamba in mitigating CI. The proposed model achieves the highest mF1 scores of 92.25% and 89.88%, along with the lowest CV values of 0.0445 and 0.0574, respectively, outperforming state-of-the-art methods. These innovations underscore the potential of HRUMamba in advancing high-resolution land cover classification in imbalanced datasets. Full article

While deep neural networks demonstrate robust performance in visual tasks, the long-tail distribution of real-world data leads to significant recognition accuracy degradation in critical scenarios such as medical human–robot affective interaction, particularly the misidentification of low-frequency negative emotions (e.g., fear and disgust) that may trigger psychological resistance in patients. Here, we propose a method based on dynamic intra-class clustering (DICC) to optimize the class imbalance problem in facial expression recognition tasks. The DICC method dynamically adjusts the distribution of majority classes by clustering them into subclasses and generating pseudo-labels, which helps the model learn more discriminative features and improve classification accuracy. By comparing with existing methods, we demonstrate that the DICC method can help the model achieve superior performance across various facial expression datasets. In this study, we conducted an in-depth evaluation of the DICC method against baseline methods using the FER2013, MMAFEDB, and Emotion-Domestic datasets, achieving improvements in classification accuracy of 1.73%, 1.97%, and 5.48%, respectively. This indicates that the DICC method can effectively enhance classification precision, especially in the recognition of minority class samples. This approach provides a novel perspective for addressing the class imbalance challenge in facial expression recognition and offers a reference for future research and applications in related fields. Full article

Distantly Supervised Relation Extraction (DSRE) aims to automatically identify semantic relationships within large text corpora by aligning with external knowledge bases. Despite the success of current methods in automating data annotation, they introduce two main challenges: label noise and data long-tail distribution. Label noise results in inaccurate annotations, which can undermine the quality of relation extraction. The long-tail problem, on the other hand, leads to an imbalanced model that struggles to extract less frequent, long-tail relations. In this paper, we introduce a novel relation extraction framework based on multi-level hierarchical attention. This approach utilizes Graph Attention Networks (GATs) to model the hierarchical structure of the relations, capturing the semantic dependencies between relation types and generating relation embeddings that reflect the overall hierarchical framework. To improve the classification process, we incorporate a multi-level classification structure guided by hierarchical attention, which enhances the accuracy of both head and tail relation extraction. A local probability constraint is introduced to ensure coherence across the classification levels, fostering knowledge transfer from frequent to less frequent relations. Experimental evaluations on the New York Times (NYT) dataset demonstrate that our method outperforms existing baselines, particularly in the context of long-tail relation extraction, offering a comprehensive solution to the challenges of DSRE. Full article

Heavy metal pollution in groundwater and the environment poses a serious threat to ecosystems and human health. In particular, heavy metal ions, such as copper (Cu), zinc (Zn) and manganese (Mn), in the leachate of metal mine tailings ponds have attracted much attention due to their high toxicity and bioaccumulation. In order to solve the problem of heavy metal pollution in groundwater caused by leachate from tailings pond of a polymetallic mine, carbon/hydroxyapatite (CHAP) prepared from animal bones was used as the medium material to systematically study its removal effect on heavy metal ions in water under static and dynamic conditions. The static experiment results showed that CHAP had excellent adsorption properties for copper, zinc, manganese and mixed ions, and the adsorption capacities were up to 80 mg/g, 67.86 mg/g and 49.29 mg/g, respectively. Dynamic experiments further confirmed the application potential of CHAP as a Permeable Reactive Barrier (PRB) medium material, which can effectively remove heavy metal ions from flowing water, having a long service life. This study provides a theoretical basis and experimental reference for the in situ remediation of heavy metal-contaminated groundwater and shows the application prospect of CHAP in the field of environmental remediation. Full article

With the rapid development of deep learning, significant progress has been made in remote sensing image target detection. However, methods based on deep learning are confronted with several challenges: (1) the inherent limitations of activation functions and downsampling operations in convolutional networks lead to frequency deviations and loss of local detail information, affecting fine-grained object recognition; (2) class imbalance and long-tail distributions further degrade the performance of minority categories; (3) large intra-class variations and small inter-class differences make it difficult for traditional deep learning methods to effectively extract fine-grained discriminative features. To address these issues, we propose a novel remote sensing aircraft recognition method. First, to mitigate the loss of local detail information, we introduce a learnable Gabor filter-based texture feature extractor, which enhances the discriminative feature representation of aircraft categories by capturing detailed texture information. Second, to tackle the long-tail distribution problem, we design a dynamic feature hallucination module that synthesizes diverse hallucinated samples, thereby improving the feature diversity of tail categories. Finally, to handle the challenge of large intra-class variations and small inter-class differences, we propose a contrastive learning module to enhance the spatial discriminative features of the targets. Extensive experiments on the large-scale fine-grained datasets FAIR1M and MAR20 demonstrate the effectiveness of our method, achieving detection accuracies of 53.56% and 89.72%, respectively, and surpassing state-of-the-art performance. The experimental results validate that our approach effectively addresses the key challenges in remote sensing aircraft recognition. Full article

Microplastic pollution is a global concern arising from the extensive production and use of plastics. The prevalence of microplastics (MPs) in the environment is escalating due in large part to the excessive use of plastics in various human-related activities. Consequently, animals are being exposed to MPs through dietary intake, which poses significant health risks to the wild populations. The objective of the study was to assess the concentration of MPs in the feces of wild long-tailed macaques (Macaca fascicularis) in the Kosumpee Forest Park (KFP) located in Northeast Thailand. KFP is situated in close proximity to the town of Kosum Phisai and experiences considerable human–primate interaction. Fresh fecal drops from 50 adult macaques were collected and sampled. MP presence in the feces was measured using density separation through visual identification under a stereomicroscope. We found a total of 396 MP particles in the feces with an average of 7.9 particles/macaque. Two forms of MPs were found in the macaques’ feces including fibers (391 pieces; 98.73%) and asymmetric fragments (5 pieces; 1.27%), with sizes mostly ranging under 1000 µm. The most observed color of MPs was blue (152 pieces; 38.48%). This study highlights the impact of anthropogenic waste and the potential health problems that can be caused to wild animals via microplastic pollution. The results contribute to the ongoing discussions on environmental health within the One Health framework. Full article

Although approaches for the online surface detection of automotive pipelines exist, low defect area rates, small-sample and long-tailed data, and the difficulty of detection due to the variable morphology of defects are three major problems faced when using such methods. In order to solve these problems, this study combines traditional visual detection methods and deep neural network technology to propose a transfer learning multi-channel fusion decision network without significantly increasing the number of network layers or the structural complexity. Each channel of the network is designed according to the characteristics of different types of defects. Dynamic weights are assigned to achieve decision-level fusion through the use of a matrix of indicators to evaluate the performance of each channel’s recognition ability. In order to improve the detection efficiency and reduce the amount of data transmission and processing, an improved ROI detection algorithm for surface defects is proposed. It can enable the rapid screening of target surfaces for the high-quality and rapid acquisition of surface defect images. On an automotive pipeline surface defect dataset, the detection accuracy of the multi-channel fusion decision network with transfer learning was 97.78% and its detection speed was 153.8 FPS. The experimental results indicate that the multi-channel fusion decision network could simultaneously take into account the needs for real-time detection and accuracy, synthesize the advantages of different network structures, and avoid the limitations of single-channel networks. Full article

Motivation: Chest X-ray (CXR) is a routine diagnostic X-ray examination for checking and screening various diseases. Automatically localizing and classifying diseases from CXR as a detection task is of much significance for subsequent diagnosis and treatment. Due to the fact that samples of some diseases are difficult to acquire, CXR detection datasets often present a long-tail distribution over different diseases. Objective: The detection performance of tail classes is very poor due to the limited number and diversity of samples in the training dataset and should be improved. Method: In this paper, motivated by a correspondence-based tracking system, we build a pipeline named RaTrack, leveraging a diffusion model to alleviate the tail class degradation problem by aligning the generation process of the tail to the head class. Then, the samples of rare classes are generated to extend the number and diversity of rare samples. In addition, we propose a filtering strategy to control the quality of the generated samples. Results: Extensive experiments on public datasets, Vindr-CXR and RSNA, demonstrate the effectiveness of the proposed method, especially for rare diseases. Full article