Search Results (98)

Short-form video platforms such as TikTok (TikTok Pte. Ltd., Singapore) host large volumes of user-generated, often ephemeral, content related to irregular migration, where relevant cues are distributed across visual scenes, on-screen text, and multilingual captions. Automatically identifying migration-related videos is challenging due to this multimodal complexity and the scarcity of labeled data in sensitive domains. This paper presents an interpretable multimodal classification framework designed for deployment under data-scarce conditions. We extract features from platform metadata, automated video analysis (Google Cloud Video Intelligence), and Optical Character Recognition (OCR) text, and compare text-only, OCR-only, and vision-only baselines against a multimodal fusion approach using Logistic Regression, Random Forest, and XGBoost. In this pilot study, multimodal fusion consistently improves class separation over single-modality models, achieving an F1-score of 0.92 for the migration-related class under stratified cross-validation. Given the limited sample size, these results are interpreted as evidence of feature separability rather than definitive generalization. Feature importance and SHAP analyses identify OCR-derived keywords, maritime cues, and regional indicators as the most influential predictors. To assess robustness under data scarcity, we apply SMOTE to synthetically expand the training set to 500 samples and evaluate performance on a small held-out set of real videos, observing stable results that further support feature-level robustness. Finally, we demonstrate scalability by constructing a weakly labeled corpus of 600 videos using the identified multimodal cues, highlighting the suitability of the proposed feature set for weakly supervised monitoring at scale. Overall, this work serves as a methodological blueprint for building interpretable multimodal monitoring pipelines in sensitive, low-resource settings. Full article

The accurate remote sensing retrieval of chlorophyll-a (Chla) concentrations in highly turbid estuarine waters remains challenging due to complex optical conditions. In this study, a small sample machine learning-based retrieval framework tailored for limited training samples was developed for the Pearl River Estuary (PRE) by integrating Sentinel-3 OLCI satellite imagery with long-term fixed-station Chla observations from the Hong Kong Environmental Protection Department. Normalized remote sensing reflectance features derived from multiple OLCI spectral bands were used as model inputs, and the performance of support vector regression (SVR) and a back propagation neural network (BPNN) was evaluated and compared with those of traditional second-order polynomial models. The results show that SVR achieves the best overall performance on both training and independent testing datasets, with a higher accuracy, smaller systematic bias, and more stable generalization capability, demonstrating its effectiveness in capturing complex nonlinear relationships under limited sample conditions. Specifically, for the training and testing datasets, the correlation coefficients between SVR-predicted and measured Chla reach 0.88 and 0.78, RMSEs are 1.75 and 1.23 mg/m³, and biases are −0.29 and 0 mg/m³, respectively. The retrieval results further reveal the clear spatiotemporal patterns of Chla concentration in the PRE, characterized by a west–high and east–low spatial distribution and pronounced seasonal migration. Elevated Chla concentrations occur mainly in the lower estuary during summer, retreat toward the upper estuary in winter, and shift to the middle estuary during spring and autumn. This study provides a practical methodological reference for the operational remote sensing monitoring of water quality in optically complex and highly turbid estuarine environments. Full article

Accurate cropland mapping in data-scarce regions remains challenging due to limited field data, strong interannual climatic variability, and heterogeneous cropping systems. This study proposes an NDVI-based training sample migration framework that transfers labeled samples from reference years in irrigated and rainfed agricultural systems to a target year using time-series similarity analysis. Ten similarity metrics representing geometric, temporal alignment, and correlation-based families were systematically evaluated to identify optimal thresholds and robust hybrid combinations for stable cropland transfer. The migrated samples were used to train a Random Forest classifier to generate binary cropland maps for 2021. Independent validation yielded overall accuracies of 86% in Kazakhstan and 95% in Uzbekistan. Comparisons with global cropland products (WorldCereal 2021 and WorldCover 2021) demonstrated improved spatial coherence and reduced misclassification, particularly in semi-arid environments. The proposed framework extends the temporal utility of existing labeled datasets and supports scalable cropland mapping without the need for repeated annual field surveys. Full article

To address the challenge of bearing load monitoring in elastically supported marine shafting systems, a multi-data fusion-based prediction model is constructed. In view of the small-sample nature of measured bearing load data, transfer learning is adopted to migrate the physical relationships embedded in finite element simulations to the measurement domain. A limited number of actual samples are used to correct the simulation data, forming a high-fidelity hybrid training set. The system—supported by air-spring isolators mounted on the raft—is divided into multiple sub-regions according to their spatial layout, establishing local mappings from air-spring pressure variations to bearing load increments to reduce model complexity. On this basis, a Stacking ensemble learning framework is further incorporated into the prediction model to integrate multi-source information such as air-spring pressure and raft strain, thereby enriching the model’s information acquisition and improving prediction accuracy. Experimental results show that the proposed transfer learning-based multi-sub-region bearing load prediction model performs significantly better than the full-parameter input model. Furthermore, the strain-enhanced Stacking-based multi-data fusion bearing load prediction model improves the characterization of shafting features and reduces the maximum prediction error. The proposed multi-data fusion modeling strategy offers a viable approach for condition monitoring and intelligent maintenance of marine shafting systems. Full article

Federated learning with data free knowledge distillation enables effective and privacy-preserving knowledge aggregation by employing generators to produce local pseudo samples during client-side model migration. However, in practical applications, data distributions across different institutions are often non-independent and identically distributed (Non-IID), which introduces bias in local models and consequently impedes the effective transfer of knowledge to the global model. In addition, insufficient local training can further exacerbate model bias, undermining overall performance. To address these challenges, we propose a heterogeneous federated learning framework that enhances knowledge transfer through guidance from global proxy data. Specifically, a noise filter is incorporated into the training of local generators to mitigate the negative impact of low-quality pseudo proxy samples on local knowledge distillation. Furthermore, a global generator is introduced to produce global pseudo proxy samples, which, together with local pseudo proxy data, are used to construct a cross attention matrix. This design effectively alleviates overfitting and underfitting issues in local models caused by data heterogeneity. Extensive experiments on publicly available datasets with heterogeneous data distributions demonstrate the superiority of the proposed framework. Results show that when the Dirichlet distribution coefficient is 0.05, our method achieves an average accuracy improvement of 5.77% over popular baselines; when the coefficient is 0.1, the improvement reaches 6.54%. Even under uniformly distributed sample classes, our model still achieves an average accuracy improvement of 7.07% compared to other methods. Full article

The aim of this work is to study the correlations of the elemental composition in the “soil–grape–wine” chain to determine the regional origin of Chardonnay grapes and wine. Soil samples (n = 40) from five vineyards in the Anapa region, Russia, taken from eight different depths, grapes from these vineyards (n = 75), and wines obtained from these grapes (n = 5) were analyzed using inductively coupled plasma atomic emission spectrometry and inductively coupled plasma mass spectrometry. The mineralogical composition of the soils was determined using thermal and X-ray phase analysis. The mineralogical composition of vineyard soils mainly consists of calcite, quartz, nontronite, vermiculite, and muscovite. According to spectrometric analysis, the distribution of both the total content and the mobile forms of elements in soil profiles turned out to be similar. The content of Na, Ca, and Sr increased with increasing sampling depth, while the content of Co, Cu, Fe, Ni, Mn, Pb, and Zn decreased. Regardless of the area of cultivation, the predominant elements in grapes are K, Ca, Na, and Mg. It is established that the elemental profiles of grapes and wine are correlated. At the same time, during the winemaking process, a decrease in the concentration of most elements (Al, Ba, Ca, Cu, K, Mg, Mn, Ni, Rb, Sr, Ti, and Zn) is observed. It has been shown that the vine is able to accumulate not only mobile but also less bioavailable forms of metals from the soil (Cu, Fe, K, Rb, Ti, and Zn), while the migration of Ca and Na remains low (<7%). Using discriminant analysis, a model of grape identification based on the concentrations of Al, Li, Mn, Na, Pb, and Rb was developed. This model demonstrated a high accuracy (100% for training and test datasets) in grape classification by region, confirming that the elemental “fingerprint” is a reliable marker of terroir. Full article

Data-driven fault location methods based on deep learning offer strong feature learning and nonlinear mapping capabilities; however, in low-voltage distribution grids (LVDG) the scarcity of high-rate sampling devices and the variability introduced by distributed renewable generation lead to data insufficiency and data imbalance, which reduce the accuracy of deep-learning-based fault location. To address this, this paper proposes an adaptive working condition-based fault location method that integrates S-transform-enhanced feature extraction with progressive transfer learning. The method clusters working conditions using k-means on a 21-dimensional indicator set covering load, photovoltaic, and voltage. For each condition, a CNN is trained on the corresponding data, and the S-transform extracts distinctive time-frequency signatures from limited measurements to separate fault points at similar distances from the feeder head. Then, progressive transfer learning with Euclidean distance-based domain adaptation migrates effective parameters from data-rich conditions to data-scarce ones through fine-tuning and medium-tuning, thereby addressing the degradation of fault-location accuracy in scenarios with limited data. Experimental validation on a 400 V LVDG demonstrates superior performance, achieving 99.80% fault location accuracy and 99.72% fault type classification. The S-transform enhancement improves fault location by 6.63%, while transfer learning maintains 96% accuracy in edge conditions using only 200 samples. Full article

Coal texture is an important factor in optimizing the characterization of coalbed methane (CBM) reservoirs, directly affecting key reservoir properties such as permeability, gas content, and production potential. This study develops an advanced methodology for coal texture classification in the Zhengzhuang Field of the Qinshui Basin, utilizing well-log data from 86 wells. Initially, 13 geophysical logging parameters were used to characterize the coal seams, resulting in a dataset comprising 2992 data points categorized into Undeformed Coal (UC), Cataclastic Coal (CC), and Granulated Coal (GC) types. After optimizing and refining the data, the dataset was reduced to 8 parameters, then further narrowed to 5 key features for model evaluation. Two primary scenarios were investigated: Scenario 1 included all 8 parameters, while Scenario 2 focused on the 5 most influential features. Five machine learning classifiers Extra Trees, Gradient Boosting, Support Vector Classifier (SVC), Random Forest, and k-Nearest Neighbors (kNN) were applied to classify coal textures. The Extra Trees classifier outperformed all other models, achieving the highest performance across both scenarios. Its peak performance was observed when 20% of the data was used for the test set and 80% for training, where it achieved a Macro F1 Score of 0.998. These findings demonstrate the potential of machine learning for improving coal texture prediction, offering valuable insights into reservoir characterization and enhancing the understanding of gas migration and accumulation processes. This methodology has significant implications for optimizing CBM resource evaluation and extraction strategies, especially in regions with limited sampling availability. Full article

In this study, we investigate the phase transition process during high-alumina, low-lithium glass-ceramics (ZnO-MgO-Li₂O-SiO₂-Al₂O₃) crystallization. The differential scanning calorimetry and high-temperature X-ray diffraction results show that approximately 10 wt.% of (Zn, Mg)Al₂O₄ crystals precipitated when the heat treatment temperature reached 850 °C, indicating that a large number of nuclei had already formed during the earlier stages of heat treatment. Field emission transmission electron microscopy used to observe the microstructure of glass-ceramics after staged heat treatment revealed that cation migration occurred during the nucleation process. Zn and Mg aggregated around Al to form (Zn, Mg)Al₂O₄ nuclei, which provided sites for crystal growth. Moreover, high-valence Zr aggregated outside the glass network, leading to the formation of nanocrystals. Raman spectroscopy analysis of samples at different stages of crystallization revealed that during spinel precipitation, the Q³ and Q⁴ structural units in the glass network increased significantly, along with an increase in the number of bridging oxygens. Highly coordinated Al originally present in the network mainly participated in spinel nucleation, effectively suppressing the subsequent formation of Li_xAl_xSi_1−xO₂, which eventually resulted in the successful preparation of glass-ceramics with (Zn, Mg)Al₂O₄ and ZrO₂ as the main crystalline phases. The grains in this glass-ceramic are all nanocrystals. Its Vickers hardness and flexural strength can reach up to 875 Hv and 350 MPa, respectively, while the visible light transmittance of the glass-ceramic reaches 81.5%. This material shows potential for applications in touchscreen protection, aircraft and high-speed train windshields, and related fields. Full article

The accurate mapping of crop planting patterns is vital for sustainable agriculture and food security, particularly in regions with complex cropping systems and limited cloud-free observations. This research focuses on the Jianghan Plain in southern China, where diverse planting structures and persistent cloud cover make consistent monitoring challenging. We integrated multi-temporal Sentinel-2 and Landsat-8 imagery from 2017 to 2021 on the Google Earth Engine platform and applied a sample migration strategy to construct multi-year training data. A random forest classifier was used to identify nine major planting patterns at a 10 m resolution. The classification achieved an average overall accuracy of 88.3%, with annual Kappa coefficients ranging from 0.81 to 0.88. A spatial analysis revealed that single rice was the dominant pattern, covering more than 60% of the area. Temporal variations in cropping patterns were categorized into four frequency levels (0, 1, 2, and 3 changes), with more dynamic transitions concentrated in the central-western and northern subregions. A multiscale geographically weighted regression (MGWR) model revealed that economic and production-related factors had strong positive associations with crop planting patterns, while natural factors showed relatively weaker explanatory power. This research presents a scalable method for mapping fine-resolution crop patterns in complex agroecosystems, providing quantitative support for regional land-use optimization and the development of agricultural policies. Full article

Fractures play a critical role in the storage and migration of hydrocarbons within subsurface reservoirs, and their characteristics can be effectively studied through core sample analysis. This study proposes an automated fracture extraction method for full-diameter core Computed Tomography (CT) images based on a deep learning framework. A semantic segmentation network called SCTNet is employed to perform high-precision semantic segmentation, while a sliding window strategy is introduced to address the challenges associated with large-scale image processing during training and inference. The proposed method achieves a mean Intersection over Union (mIoU) of 72.14% and a pixel-level segmentation accuracy of 97% on the test dataset, outperforming traditional thresholding techniques and several state-of-the-art deep learning models. Besides fracture detection, the method enables quantitative characterization of fracture-related parameters, including fracture proportion, dip angle, strike, and aperture. Experimental results indicate that the proposed approach provides a reliable and efficient solution for the interpretation of large-volume CT data. Compared to manual evaluation, the method significantly accelerates the analysis process—reducing time from hours to minutes—and demonstrates strong potential to enhance intelligent workflows for geological core fracture analysis. Full article

Machine learning (ML) techniques are becoming increasingly valuable for modeling the transport of pollutants in plant systems. However, two challenges (small sample sizes and a lack of quantitative calculation functions) remain when using ML to predict migration in hydroponic systems. For the bioaccumulation of per- and polyfluoroalkyl substances, we studied the key factors and quantitative calculation equations based on data augmentation, ML, and symbolic regression. First, feature expansion was performed on the input data after data preprocessing; the most important step was data augmentation. The original training set was expanded nine times by combining the synthetic minority oversampling technique and a variational autoencoder. Subsequently, the four ML models were applied to the test set to predict the selected output parameters. Categorical boosting (CatBoost) had the highest prediction accuracy (R² = 0.83). The Shapley Additive Explanation values indicated that molecular weight and exposure time were the most important parameters. We applied three symbolic regression models to obtain accurate prediction equations based on the original and augmented data. Based on augmented data, the high-dimensional sparse interaction equation exhibited the highest accuracy (R² = 0.776). Our results indicate that this method could provide crucial insights into absorption and accumulation in plant roots. Full article

To address the challenges in identifying NAPL contamination within low-permeability clay sites, this study innovatively integrates high-density electrical resistivity tomography (ERT) with a UNet deep learning model to establish an intelligent contamination detection system. Taking an industrial site in Shanghai as the research object, we collected apparent resistivity data using the WGMD-9 system, obtained resistivity profiles through inversion imaging, and constructed training sets by generating contamination labels via K-means clustering. A semantic segmentation model with skip connections and multi-scale feature fusion was developed based on the UNet architecture to achieve automatic identification of contaminated areas. Experimental results demonstrate that the model achieves a mean Intersection over Union (mIoU) of 86.58%, an accuracy (Acc) of 99.42%, a precision (Pre) of 75.72%, a recall (Rec) of 76.80%, and an F1 score (f1) of 76.23%, effectively overcoming the noise interference in electrical anomaly interpretation through conventional geophysical methods in low-permeability clay, while outperforming DeepLabV3, DeepLabV3+, PSPNet, and LinkNet models. Time-lapse resistivity imaging verifies the feasibility of dynamic monitoring for contaminant migration, while the integration of the VGG-16 encoder and hyperparameter optimization (learning rate of 0.0001 and batch size of 8) significantly enhances model performance. Case visualization reveals high consistency between segmentation results and actual contamination distribution, enabling precise localization of spatial morphology for contamination plumes. This technological breakthrough overcomes the high-cost and low-efficiency limitations of traditional borehole sampling, providing a high-precision, non-destructive intelligent detection solution for contaminated site remediation. Full article

Under laboratory conditions for recording a small amount of data, the characteristics of the phenomena distribution become a limitation of machine learning and advanced deep learning concepts for the diagnosis and localization of mechanical wear faults. In this paper, we adopt the combination of the diffusion model and TTT (test-time training), based on the sample distribution of feature data under the laboratory conditions, and we use the pre-trained decoder to decode the data into a continuous potential representation of natural language for sampling, to achieve data regeneration. Subsequently, the TTT algorithm becomes a model with weights in the hidden state itself. The gradient step on the self-supervised loss is selected as the update rule, which is trained synchronously during the testing time, adhering to the concept of migration learning, to construct a high-dimensional mapping relationship between the feature parameters and the failure modes of the mechanical wear. The final validation results show that the diagnosis accuracy reaches more than 95% for six types of typical aero-engine mechanical wear faults. Full article

Seismic interferometry using ambient noise provides an effective approach for subsurface imaging through reconstructing passive virtual source (PVS) responses. Traditional crosscorrelation (CC) seismic interferometry relies on a uniform dense distribution of passive sources in the subsurface, which is often challenging in practice. The multidimensional deconvolution method (MDD) alleviates reliance on passive-source distribution, but requires wavefield decomposition of the original data. This is difficult to accurately achieve for uncorrelated noise sources, leading to the existence of non-physical artifacts in the reconstructed PVS data. To address this issue, this study proposes a method to improve the accuracy of PVS data reconstruction using an enhanced U-Net. This data-driven approach circumvents the challenge of noise wavefield decomposition encountered in the traditional MDD. By integrating a feature fusion module into U-Net, multi-scale sampling information is leveraged to improve the network’s ability to capture detailed PVS data features. The combination of active-source data constraints and the modified MDD further optimizes PVS data retrieval during training. Numerical tests show that the proposed method effectively recovers waveform information in PVS retrieval records with non-ideally distributed sources, suppressing coherent noise and false events. The reconstructed recordings have a clear advantage in the reverse time migration (RTM) imaging results, with strong generalization performance across various velocity models. Full article