Search Results (1,504)

This study assesses the suitability of UAV aerial imagery-based photogrammetry for reconstructing underwater riverbed topography and its application in two-dimensional (2D) hydrodynamic modelling, with a particular focus on comparing RGB, multispectral, and fused RGB–multispectral imagery. Four Lithuanian rivers—Verknė, Šušvė, Jūra, and Mūša—were selected to represent a wide range of hydromorphological and hydraulic conditions, including variations in bed texture, vegetation cover, and channel complexity. High-resolution digital elevation models (DEMs) were generated from field-based surveys and UAV imagery processed using Structure-from-Motion photogrammetry. Two-dimensional hydrodynamic models were created and calibrated in HEC-RAS 6.5 using measurement-based DEMs and subsequently applied using photogrammetry-derived DEMs to isolate the influence of terrain input on model performance. The results showed that UAV-derived DEMs systematically overestimate riverbed elevation, particularly in deeper or vegetated sections, resulting in underestimated water depths. RGB imagery provided greater spatial detail but was more susceptible to local anomalies, whereas multispectral imagery produced smoother surfaces with a stronger positive elevation bias. The fusion of RGB and multispectral imagery consistently reduced spatial noise and improved hydrodynamic simulation performance across all river types. Despite moderate vertical deviations of 0.10–0.25 m, relative flow patterns and velocity distributions were reproduced with acceptable accuracy. The findings demonstrate that combined spectral UAV aerial imagery in photogrammetry is a robust and cost-effective alternative for hydrodynamic modelling in shallow lowland rivers, particularly where relative hydraulic characteristics are of primary interest. Full article

This study proposes 4D-MRSimNet, a framework that employs attention mechanisms to focus on distinct dimensions. The approach applies enhancements to key responses in the spatial and spectral domains and provides a characterization of dynamic evolution in temporal domain, which extracts and integrates complementary emotional features to facilitate final classification. At the feature level, differential entropy (DE) and power spectral density (PSD) are combined within four core frequency bands ($\theta$, $\alpha$, $\beta$, and $\gamma$). These bands are recognized as closely related to emotional processing. This integration constructs a complementary feature representation that preserves both energy distribution and entropy variability. These features are organized into a 4D representation that integrates electrode topology, frequency characteristics, and temporal dependencies inherent in EEG signals. At the network level, a multi-resolution convolutional module embedded with SimAM attention extracts spatial and spectral features at different scales and adaptively emphasizes key information. A bidirectional GRU (BiGRU) integrated with temporal attention further emphasizes critical time segments and strengthens the modeling of temporal dependencies. Experiments show that our method achieves an accuracy of 97.68% for valence and 97.61% for arousal on the DEAP dataset and 99.60% for valence and 99.46% for arousal on the DREAMER dataset. The results demonstrate the effectiveness of complementary feature fusion, multidimensional feature representation, and the complementary dual attention enhancement strategy for EEG emotion recognition. Full article

Xinfeng County, Shaoguan City, Guangdong Province, China, is a typical vegetation-covered area that suffers from severe attenuation of rock and mineral spectral information in remote sensing images owing to dense vegetation. This situation limits the accuracy of traditional lithological mapping methods, making them unable to meet geological mapping demands under complex conditions, and thus necessitating a tailored lithological identification model. To address this issue, in this study, the penetration capability of microwave remote sensing (for extracting indirect textural features of lithology) was combined with the spectral superiority of hyperspectral remote sensing (for capturing lithological spectral features), resulting in a dual-branch deep-learning framework for lithological classification based on multisource remote sensing data. The framework independently extracts features from Sentinel-1 imagery and Gaofen-5 data, integrating three key modules: texture feature extraction, spatial–spectral feature extraction, and attention-based adaptive feature fusion, to realize deep and efficient fusion of heterogeneous remote sensing information. Ablation and comparative experiments were conducted to evaluate each module’s contribution. The results show that the dual-branch architecture effectively captures the complementary and discriminative characteristics of multimodal data, and that the encoder–decoder structure demonstrates strong robustness under complex conditions such as dense vegetation. The final model achieved 97.24% overall accuracy and 90.43% mean intersection-over-union score, verifying its effectiveness and generalizability in complex geological environments. The proposed multi-source remote sensing–based lithological classification model overcomes the limitations of single-source data by integrating indirect lithological texture features containing vegetation structural information with spectral features, thereby providing a viable approach for lithological mapping in vegetated regions. Full article

Facial expression recognition (FER) is a key task in affective computing and human–computer interaction, aiming to decode facial muscle movements into emotional categories. Although deep learning-based FER has achieved remarkable progress, robust recognition under uncontrolled conditions (e.g., illumination change, pose variation, occlusion, and cultural diversity) remains challenging. Traditional Convolutional Neural Networks (CNNs) are effective at local feature extraction but limited in modeling global dependencies, while Vision Transformers (ViT) provide global context modeling yet often neglect fine-grained texture and frequency cues that are critical for subtle expression discrimination. Moreover, existing approaches usually focus on single-domain representations and lack adaptive strategies to integrate heterogeneous cues across spatial, semantic, and spectral domains, leading to limited cross-domain generalization. To address these limitations, this study proposes a unified Multi-Domain Feature Enhancement and Fusion (MDFEFT) framework that combines a ViT-based global encoder with three complementary branches—channel, spatial, and frequency—for comprehensive feature learning. Taking into account the approximately bilateral symmetry of human faces and the asymmetric distortions introduced by pose, occlusion, and illumination, the proposed MDFEFT framework is designed to learn symmetry-aware and asymmetry-robust representations for facial expression recognition across diverse domains. An adaptive Cross-Domain Feature Enhancement and Fusion (CDFEF) module is further introduced to align and integrate heterogeneous features, achieving domain-consistent and illumination-robust expression understanding. The experimental results show that the proposed method consistently outperforms existing CNN-, Transformer-, and ensemble-based models. The proposed model achieves accuracies of 0.997, 0.796, and 0.776 on KDEF, FER2013, and RAF-DB, respectively. Compared with the strongest baselines, it further improves accuracy by 0.3%, 2.2%, and 1.9%, while also providing higher F1-scores and better robustness in cross-domain testing. These results confirm the effectiveness and strong generalization ability of the proposed framework for real-world facial expression recognition. Full article

The mineral deposits in Namibia’s Erongo region are renowned and frequently associated with complex geological environments, including calcrete-hosted paleochannels and hydrothermal alteration zones. Mineral extraction is hindered by high operational costs, restricted accessibility and stringent environmental regulations. To address these challenges, this study proposes an integrated approach that combines satellite remote sensing and machine learning to map and identify mineralisation-indicative zones. Sentinel 2 Multispectral Instrument (MSI) and Landsat 8 Operational Land Imager (OLI) multispectral data were employed due to their global coverage, spectral fidelity and suitability for geological investigations. Normalized Difference Vegetation Index (NDVI) masking was applied to minimise vegetation interference. Spectral indices—the Clay Index, Carbonate Index, Iron Oxide Index and Ferrous Iron Index—were developed and enhanced using false-colour composites. Principal Component Analysis (PCA) was used to reduce redundancy and extract significant spectral patterns. Supervised classification was performed using Support Vector Machine (SVM), Random Forest (RF) and Maximum Likelihood Classification (MLC), with validation through confusion matrices and metrics such as Overall Accuracy, User’s Accuracy, Producer’s Accuracy and the Kappa coefficient. The results showed that RF achieved the highest accuracy on Landsat 8 and MLC outperformed others on Sentinel 2, while SVM showed balanced performance. Sentinel 2’s higher spatial resolution enabled improved delineation of alteration zones. This approach supports efficient and low-impact mineral prospecting in remote environments. Full article

Convolutional Neural Networks (CNNs) have been extensively applied for the extraction of deep features in hyperspectral imagery tasks. However, traditional 3D-CNNs are limited by their fixed-size receptive fields and inherent locality. This restricts their ability to capture multi-scale objects and model long-range dependencies, ultimately hindering the representation of large-area land-cover structures. To overcome these drawbacks, we present a new framework designed to integrate multi-scale feature fusion and a hierarchical attention mechanism for hyperspectral image classification. Channel-wise Squeeze-and-Excitation (SE) and Convolutional Block Attention Module (CBAM) spatial attention are combined to enhance feature representation from both spectral bands and spatial locations, allowing the network to emphasize critical wavelengths and salient spatial structures. Finally, by integrating the self-attention inherent in the Transformer architecture with a Cross-Attention Fusion (CAF) mechanism, a local-global feature fusion module is developed. This module effectively captures extended-span interdependencies present in hyperspectral remote sensing images, and this process facilitates the effective integration of both localized and holistic attributes. On the Salinas Valley dataset, the proposed method delivers an Overall Accuracy (OA) of 0.9929 and an Average Accuracy (AA) of 0.9949, attaining perfect recognition accuracy for certain classes. The proposed model demonstrates commendable class balance and classification stability. Across multiple publicly available hyperspectral remote sensing image datasets, it systematically produces classification outcomes that significantly outperform those of established benchmark methods, exhibiting distinct advantages in feature representation, structural modeling, and the discrimination of complex ground objects. Full article

Hyperspectral images (HSIs) have been broadly applied in remote sensing, environmental monitoring, agriculture, and other fields due to their rich spectral information and complex spatial properties. However, the inherent redundancy, spectral aliasing, and spatial heterogeneity of high-dimensional data pose significant challenges to classification accuracy. Therefore, this study proposes STM-Net, a hybrid deep learning model that integrates SSRE (Spectral–Spatial Residual Extraction Module), Transformer, and MDRM (Multi-scale Differential Residual Module) architectures to comprehensively exploit spectral–spatial features and enhance classification performance. First, the SSRE module employs 3D convolutional layers combined with residual connections to extract multi-scale spectral–spatial features, thereby improving the representation of both local and deep-level characteristics. Second, the MDRM incorporates multi-scale differential convolution and the Convolutional Block Attention Module mechanism to refine local feature extraction and enhance inter-class discriminability at category boundaries. Finally, the Transformer branch equipped with a Dual-Branch Global-Local (DBGL) mechanism integrates local convolutional attention and global self-attention, enabling synergistic optimization of long-range dependency modeling and local feature enhancement. In this study, STM-Net is extensively evaluated on three benchmark HSI datasets: Indian Pines, Pavia University, and Salinas. Additionally, experimental results demonstrate that the proposed model consistently outperforms existing methods regarding OA, AA, and the Kappa coefficient, exhibiting superior generalization capability and stability. Furthermore, ablation studies validate that the SSRE, MDRM, and Transformer components each contribute significantly to improving classification performance. This study presents an effective spectral–spatial feature fusion framework for hyperspectral image classification, offering a novel technical solution for remote sensing data analysis. Full article

Flexible grid regulation necessitates Francis turbines to operate at heads of 120–180 m (compared to the rated head of 154.6 m), breaking the designed rotational symmetry and inducing hydraulic instabilities that threaten structural integrity and operational reliability. This study presents extensive field measurements of pressure pulsations in a 600 MW prototype Francis turbine operating at heads of 120–180 m and loads of 20–600 MW across 77 operating conditions (7 head levels × 11 load points). We strategically positioned high-precision piezoelectric pressure sensors at three critical locations—volute inlet, vaneless space, and draft tube cone—to capture the amplitude and frequency characteristics of symmetry-breaking phenomena. Advanced signal processing revealed three distinct mechanisms with characteristic pressure pulsation signatures: (1) Draft tube rotating vortex rope (RVR) represents spontaneous breaking of axial symmetry, exhibiting helical precession at 0.38 Hz (approximately 0.18 f_n, where f_n = 2.08 Hz) with maximum peak-to-peak amplitudes of 108 kPa (87% of the rated pressure p_rated = 124 kPa) at H = 180 m and P = 300 MW, demonstrating approximately 70% amplitude reduction potential through load-based operational strategies. (2) Vaneless space rotor-stator interaction (RSI) reflects periodic disruption of the combined C₂₄ × C₁₃ symmetry at the blade-passing frequency of 27.1 Hz (N_r × f_n = 13 × 2.08 Hz), reaching peak amplitudes of 164 kPa (132% p_rated) at H = 180 m and P = 150 MW, representing the most severe symmetry-breaking phenomenon. (3) Volute multi-point excitation exhibits broadband spectral characteristics (4–10 Hz) with peak amplitudes of 146 kPa (118% p_rated) under small guide vane openings. The spatial amplitude hierarchy—vaneless space (164 kPa) > volute (146 kPa) > draft tube (108 kPa)—directly correlates with the local symmetry-breaking intensity, providing quantitative evidence for the relationship between geometric symmetry disruption and hydraulic excitation magnitude. Systematic head-dependent amplitude increases of 22–43% across all monitoring locations are attributed to effects related to Euler head scaling and Reynolds number variation, with the vaneless space demonstrating the highest sensitivity (0.83 kPa/m, equivalent to 0.67% p_rated/m). The study establishes data-driven operational guidelines identifying forbidden operating regions (H = 160–180 m, P = 20–150 MW for vaneless space; H = 160–180 m, P = 250–350 MW for draft tube) and critical monitoring frequencies (0.38 Hz for RVR, 27.1 Hz for RSI), providing essential reference data for condition monitoring systems and operational optimization of large Francis turbines functioning as flexible grid-regulating units in renewable energy integration scenarios. Full article

Accurate prediction of air pollutant concentrations, particularly fine particulate matter (PM_2.5), is essential for controlling and preventing heavy pollution incidents by providing early warnings of harmful substances in the atmosphere. This study proposes a novel spatiotemporal model for PM_2.5 concentration prediction based on a Conditional Wasserstein Generative Adversarial Network with Gradient Penalty (CWGAN-GP). The framework incorporates three key design components: First, the generator employs an Inception-style Convolutional Long Short-Term Memory (ConvLSTM) network, integrating parallel multi-scale convolutions and hierarchical normalization. This design enhances multi-scale spatiotemporal feature extraction while effectively suppressing boundary artifacts via a map-masking layer. Second, the discriminator adopts an architecturally enhanced U-Net, incorporating spectral normalization and shallow instance normalization. Feature-guided masked skip connections are introduced, and the output is designed as a raw score map to mitigate premature saturation during training. Third, a composite loss function is utilized, combining adversarial loss, feature-matching loss, and inter-frame spatiotemporal smoothness. A sliding-window conditioning mechanism is also implemented, leveraging multi-level features from the discriminator for joint spatiotemporal optimization. Experiments conducted on multi-source gridded data from Dongguan demonstrate that the model achieves a 12 h prediction performance with a Root Mean Square Error (RMSE) of 4.61 μg/m³, a Mean Absolute Error (MAE) of 6.42 μg/m³, and a Coefficient of Determination (R²) of 0.80. The model significantly alleviates performance degradation in long-term predictions when the forecast horizon is extended from 3 to 12 h, the RMSE increases by only 1.84 μg/m³, and regional deviations remain within ±3 μg/m³. These results indicate strong capabilities in spatial topology reconstruction and robustness against concentration anomalies, highlighting the model’s potential for hyperlocal air quality early warning. It should be noted that the empirical validation is limited to the specific environmental conditions of Dongguan, and the model’s generalizability to other geographical and climatic settings requires further investigation. Full article

Early detection of breast cancer commonly relies on imaging technologies such as ultrasound, mammography and MRI. Among these, breast ultrasound is widely used by radiologists to identify and assess lesions. In this study, we developed image segmentation techniques and multiclass classification artificial intelligence (AI) tools based on pretrained models to segment lesions and detect breast cancer. The proposed workflow includes both the development of segmentation models and development of a series of classification models to classify ultrasound images as normal, benign or malignant. The pretrained models were trained and evaluated on the Breast Ultrasound Images (BUSI) dataset, a publicly available collection of grayscale breast ultrasound images with corresponding expert-annotated masks. For segmentation, images and ground-truth masks were used to pretrained encoder (ResNet18, EfficientNet-B0 and MobileNetV2)–decoder (U-Net, U-Net++ and DeepLabV3) models, including the DeepLabV3 architecture integrated with a Frequency-Domain Feature Enhancement Module (FEM). The proposed FEM improves spatial and spectral feature representations using Discrete Fourier Transform (DFT), GroupNorm, dropout regularization and adaptive fusion. For classification, each image was assigned a label (normal, benign or malignant). Optuna, an open-source software framework, was used for hyperparameter optimization and for the testing of various pretrained models to determine the best encoder–decoder segmentation architecture. Five different pretrained models (ResNet18, DenseNet121, InceptionV3, MobielNetV3 and GoogleNet) were optimized for multiclass classification. DeepLabV3 outperformed other segmentation architectures, with consistent performance across training, validation and test images, with Dice Similarity Coefficient (DSC, a metric describing the overlap between predicted and true lesion regions) values of 0.87, 0.80 and 0.83 on training, validation and test sets, respectively. ResNet18:DeepLabV3 achieved an Intersection over Union (IoU) score of 0.78 during training, while ResNet18:U-Net++ achieved the best Dice coefficient (0.83) and IoU (0.71) and area under the curve (AUC, 0.91) scores on the test (unseen) dataset when compared to other models. However, the proposed Resnet18: FrequencyAwareDeepLabV3 (FADeepLabV3) achieved a DSC of 0.85 and an IoU of 0.72 on the test dataset, demonstrating improvements over standard DeepLabV3. Notably, the frequency-domain enhancement substantially improved the AUC from 0.90 to 0.98, indicating enhanced prediction confidence and clinical reliability. For classification, ResNet18 produced an F1 score—a measure combining precision and recall—of 0.95 and an accuracy of 0.90 on the training dataset, while InceptionV3 performed best on the test dataset, with an F1 score of 0.75 and accuracy of 0.83. We demonstrate a comprehensive approach to automate the segmentation and multiclass classification of breast cancer ultrasound images into benign, malignant or normal transfer learning models on an imbalanced ultrasound image dataset. Full article

Electroencephalography (EEG)-based brain–computer interface (BCI) mimics the brain’s intrinsic information-processing mechanisms by translating neural oscillations into actionable commands. In motor imagery (MI) BCI, imagined movements evoke characteristic patterns over the sensorimotor cortex, forming a biomimetic channel through which internal motor intentions are decoded. However, this biomimetic interaction is highly vulnerable to signal degradation, particularly in mobile or low-resource environments where low sampling frequencies obscure these MI-related oscillations. To address this limitation, we propose a robust MI classification framework that integrates spatial, spectral, and temporal dynamics through a filter bank common spatial pattern with time segmentation (FBCSP-TS). This framework classifies motor imagery tasks into four classes (left hand, right hand, foot, and tongue), segments EEG signals into overlapping time domains, and extracts frequency-specific spatial features across multiple subbands. Segment-level predictions are combined via soft voting, reflecting the brain’s distributed integration of information and enhancing resilience to transient noise and localized artifacts. Experiments performed on BCI Competition IV datasets 2a (250 Hz) and 1 (100 Hz) demonstrate that FBCSP-TS outperforms CSP and FBCSP. A paired t-test confirms that accuracy at 110 Hz is not significantly different from that at 250 Hz (p < 0.05), supporting the robustness of the proposed framework. Optimal temporal parameters (window length = 3.5 s, moving length = 0.5 s) further stabilize transient-signal capture and improve SNR. External validation yielded a mean accuracy of 0.809 ± 0.092 and Cohen’s kappa of 0.619 ± 0.184, confirming strong generalizability. By preserving MI-relevant neural patterns under degraded conditions, this framework advances practical, biomimetic BCI suitable for wearable and real-world deployment. Full article

Harmful Algal Blooms (HABs) in large river systems present significant challenges for water quality monitoring, with traditional in-situ sampling methods limited by spatial and temporal coverage. This study evaluates the effectiveness of machine learning techniques applied to Landsat spectral data for detecting and quantifying HABs in the Ohio River system, with particular focus on the unprecedented 2015 bloom event. Our methodology combines Google Earth Engine (GEE) for satellite data processing with an ensemble machine learning approach incorporating Support Vector Regression (SVR), Neural Networks (NN), and Extreme Gradient Boosting (XGB). Analysis of Landsat 7 and 8 data revealed that the 2015 HAB event had both broader spatial extent (636.5 river miles) and earlier onset (5–7 days) than detected through conventional monitoring. The ensemble model achieved a correlation coefficient of 0.85 with ground-truth measurements and demonstrated robust performance in detecting varying bloom intensities (R² = 0.82). Field validation using ORSANCO monitoring stations confirmed the model’s reliability (Nash-Sutcliffe Efficiency = 0.82). The integration of multispectral indices, particularly the Floating Algae Index (FAI) and Normalized Difference Chlorophyll Index (NDCI), enhanced detection accuracy by 23% compared to single-index approaches. The GEE-based framework enables near real-time processing and automated alert generation, making it suitable for operational deployment in water management systems. These findings demonstrate the potential for satellite-based HAB monitoring to complement existing ground-based systems and establish a foundation for improved early warning capabilities in large river systems through the integration of remote sensing and machine learning techniques. Full article

Hyperspectral anomaly detection (HAD) refers to a method of identifying abnormal targets through the differences in spectral separabilities of anomaly versus background clutter. It plays a significant role in fields such as commercial agriculture, for instance, in pest and disease monitoring and environmental monitoring. Collaborative representation detector (CRD) is a classic hyperspectral anomaly detection method. However, by constructing a sliding dual window, it leads to a high computational complexity and thus takes a relatively long time. In response to the deficiencies existing in that CRD method, we propose a method that first extracts extended morphological profiles (EMP) and then uses the obtained feature images to construct K-means CRD (EMPKCRD). This method performs window reconstruction on complex hyperspectral background pixels through the K-means clustering algorithm to separate abnormal pixels with similar features and obtain the background dictionary matrix. The method leverages the observation that background pixels can be effectively approximated by a linear combination of their spatially adjacent pixels, whereas anomalous pixels, due to their distinct nature, cannot be similarly reconstructed from their local neighborhood. This fundamental disparity in reconstructibility is then exploited to separate anomalies from the background. Then, anomaly detection can be carried out on this matrix faster, avoiding the high computational complexity caused by the use of a sliding dual window. Through comparative simulation experiments with seven widely used algorithms at present on three real-world datasets, the empirical evaluations validate that this method has excellent performance while exhibiting a favorable balance between detection precision and operational speed. Full article

Background: Artificial intelligence (AI) emerges as a powerful tool to assist breast cancer screening; however, its integration into different mammographic modalities remains insufficiently explored. Digital Mammography (DM) is widely accessible but presents limitations in dense breast tissue, whereas Contrast-Enhanced Spectral Mammography (CESM) provides functional information that enhances lesion visualization. Understanding how deep learning models behave across these modalities, and determining whether their decision-making patterns remain consistent, is essential for equitable clinical adoption. Methods: This study evaluates three convolutional neural network (CNN) architectures, ResNet-18, DenseNet-121, and EfficientNet-B0, for binary classification of breast lesions using DM and CESM images from the public CDD-CESM dataset (2006 images, three diagnostic classes). The models are trained separately on DM and CESM using three classification tasks: Normal vs. Benign, Benign vs. Malignant, and Normal vs. Malignant. A 3-fold cross-validation scheme and an independent test set are employed. Training uses transfer learning with ImageNet weights, weighted binary cross-entropy (BCE) loss, and SHapley Additive exPlanations (SHAP) analysis to visualize pixel-level relevance of model decisions. Results: CESM yields higher performance in the Normal vs. Benign and Benign vs. Malignant tasks, whereas DM achieves the highest discriminative ability in the Normal vs. Malignant comparison (EfficientNet-B0: AUC = 97%, Accuracy = 93.15%), surpassing the corresponding CESM results (AUC = 93%, Accuracy = 85.66%). SHAP attribution maps reveal anatomically coherent decision patterns in both modalities, with CESM producing sharper and more localized relevance regions due to contrast uptake, while DM exhibits broader yet spatially aligned attention. Across architectures, EfficientNet-B0 demonstrates the most stable performance and interpretability. Conclusions: CESM enhances subtle lesion discrimination through functional contrast, whereas DM, despite its simpler acquisition and wider availability, provides highly accurate and explainable outcomes when combined with modern CNNs. The consistent SHAP-based relevance observed across modalities indicates that both preserve clinically meaningful information. To the best of our knowledge, this study is the first to directly compare DM and CESM under identical preprocessing, training, and evaluation conditions using explainable deep learning models. Full article

Accurate identification and monitoring of rice cultivation areas are essential for food security and sustainable agricultural development. However, regions with frequent cloud cover, high rainfall, and fragmented fields often face challenges due to the absence of temporal features caused by cloud and rain interference, as well as spectral confusion from scattered plots, which hampers rice recognition accuracy. To address these issues, this study employs a Satellite Image Time Series Transformer (SITS-Former) model, enhanced with the integration of diverse phenological features to improve rice phenology representation and enable precise rice identification. The methodology constructs a rice phenological feature set that combines temporal, spatial, and spectral information. Through its self-attention mechanism, the model effectively captures growth dynamics, while multi-scale convolutional modules help suppress interference from non-rice land covers. The study utilized Sentinel-2 satellite data to analyze rice distribution in Wuxi City. The results demonstrated an overall classification accuracy of 0.967, with the estimated planting area matching 91.74% of official statistics. Compared to traditional rice distribution analysis methods, such as Random Forest, this approach outperforms in both accuracy and detailed presentation. It effectively addresses the challenge of identifying fragmented rice fields in regions with persistent cloud cover and heavy rainfall, providing accurate mapping of cultivated areas in difficult climatic conditions while offering valuable baseline data for yield assessments. Full article