Search Results (287)

The synergistic integration of Hyperspectral Imaging (HSI) and Light Detection and Ranging (LiDAR) data has become a pivotal strategy in remote sensing for precise land-cover classification. However, existing multimodal deep learning frameworks frequently suffer from intrinsic limitations, including rigid feature extraction protocols, underutilization of LiDAR-derived textural information, and asymmetric fusion mechanisms that fail to balance the contribution of spectral and elevation features effectively. To address these challenges, this paper proposes a novel framework named DGC-BCAF, which integrates Dynamic Group Convolution and Bidirectional Guided Cross-Attention Fusion to achieve adaptive feature representation and robust cross-modal interaction. First, a Dynamic Group Convolution (DGConv) module embedded within a ResNet18 backbone is designed to function as the central spatial context extractor. Unlike traditional group convolution, this module learns a dynamic relationship matrix to automatically group input channels, thereby facilitating flexible and context-aware feature representation that adapts to complex spatial distributions. Second, to overcome the insufficient exploitation of elevation data, we introduce a dedicated LiDAR texture encoding branch. This branch innovatively fuses Gray-Level Co-occurrence Matrix (GLCM) statistical features with multi-scale convolutional representations, capturing both geometric height information and fine-grained surface textural details that are critical for distinguishing objects with similar elevations. Finally, central to our architecture is the Bidirectional Cross-Attention Fusion (BCAF) module. Unlike standard unidirectional fusion approaches, BCAF employs a LiDAR geometry to guide the selection of salient spectral bands, while simultaneously utilizing spectral signatures to emphasize informative LiDAR channels. This mutual guidance ensures a balanced contribution from both modalities. Extensive experiments conducted on three benchmark datasets—Houston 2013, Trento, and MUUFL—demonstrate that DGC-BCAF consistently outperforms state-of-the-art methods in terms of overall accuracy, average accuracy, and Kappa coefficient. The results confirm that the proposed adaptive grouping and bidirectional guidance strategies significantly improve classification performance, particularly in distinguishing spectrally similar materials and delineating complex urban structures. Full article

Filtering constitutes a critical step in the post-processing of airborne Light Detection And Ranging (LiDAR) data. Over the past decade, machine learning has emerged as a prominent methodological paradigm across numerous disciplines, attracting significant research interest in its application to LiDAR filtering. From a machine learning perspective, filtering is essentially a binary classification task that aims to discriminate between ground and non-ground points. However, the limited information inherent in point clouds often leads to the generation of highly correlated features, particularly those derived from height data, which can compromise filtering accuracy. To address this issue, feature selection becomes imperative. In this study, we employed height-based mutual information as a criterion to identify and eliminate less discriminative features for filtering. The AdaBoost (Adaptive Boosting) algorithm was adopted as the classifier for point cloud filtering. For each point, nineteen features were derived from the raw LiDAR point cloud based on height and other geometric attributes within a defined neighborhood. The performance of the proposed feature selection approach was evaluated using benchmark datasets provided by the International Society for Photogrammetry and Remote Sensing (ISPRS). Experimental results demonstrate that the method is effective and reliable. After removing three selected features, the average kappa coefficient improved, along with a reduction in three categories of error, although a slight increase in Type II error (0.15%) was observed. Full article

The grain aphid (Sitobion avenae) is a major pest of winter wheat, causing significant yield losses through direct feeding and as a vector of barley yellow dwarf virus (BYDV). Populations can increase rapidly under moderate temperatures and low rainfall, potentially leading to severe infestations if not effectively monitored and managed. This study develops and validates a UAV-based RGB imaging methodology, which relies on deep learning for accurate detection and assessment of Sitobion avenae in wheat crops. The RGB images are preliminarily filtered using “histogram equalization”, which allows for highlighting the infested areas. An experimental study was conducted under the specific climatic conditions of Southern Dobruja, Bulgaria, to quantify Sitobion avenae infestations. Three neural network architectures were used (DeepLabv3, U-Net, and PSPNet) in combination with three backbone models: ResNet34, ResNet50, and ResNet101. The optimal combination was determined to be the U-Net + ResNet101 model, which achieved an average F1 score of 0.982 and a Cohen’s Kappa coefficient of 0.966. The results demonstrate that UAV-based detection allows precise mapping of infested areas, enabling targeted insecticide applications and effective pest management while substantially reducing chemical inputs. These findings indicate that the proposed framework provides a reliable and scalable tool for precision pest monitoring and control in winter wheat. Full article

Cyanobacterial blooms pose a serious threat to freshwater ecosystems, necessitating accurate remote sensing monitoring. Although red-edge bands show potential in terrestrial monitoring, their multi-dimensional features (i.e., spectral, textural, and index-based characteristics) remain underutilized for aquatic blooms. This study leverages the dual red-edge bands (710 nm and 750 nm) of GF-6/WFV to enhance cyanobacterial bloom identification in Lake Taihu. Multi-temporal images from 2019–2023 were used to construct red-edge features in three dimensions: spectral (evaluated via adaptive band selection method) and Jeffries–Matusita–Bhattacharyya distance), texture (based on Gray Level Co-occurrence Matrix and principal component analysis), and indices (nine vegetation indices ranked by Random Forest importance). Twelve feature-combination schemes were designed and implemented with a Random Forest classifier. Results show that red-edge features consistently improve identification accuracy. Quantitatively, compared to the basic four-band (RGBN) combination, the 710 nm band improved spectral separability by an average of 9.63%, whereas the 750 nm band yielded a lower average improvement of 5.69%. Red-edge indices, especially the modified chlorophyll absorption reflectance index 1 (MCARI1) and normalized difference red-edge index (NDRE), exhibited higher importance than non-red-edge indices. All schemes incorporating red-edge features achieved mean overall accuracies of 92.8–94.9% and Kappa coefficients of 0.86–0.94, surpassing the basic four-band scheme. Among these features, red-edge indices contributed most significantly to accuracy gains, increasing the overall accuracy by an average of 0.36–6.06% and the Kappa coefficient by up to 0.06. The enhancement effect of the red-edge 710 nm band features was superior to that of the 750 nm band. This study demonstrates that multi-dimensional red-edge features effectively enhance the identification accuracy of cyanobacterial blooms and provides a methodological reference for operational GF-6 applications in water quality monitoring. Full article

Automated waste classification is essential for improving recycling efficiency and supporting sustainable waste management systems. However, conventional convolutional neural network (CNN) approaches primarily focus on localized feature extraction, which may limit their ability to capture complex spatial relationships in heterogeneous waste materials. This study proposes DeepWasteSort-SI-SSO, a Vision Transformer (ViT)-based framework enhanced with a Self-Improved Sparrow Search Optimization (SI-SSO) strategy for hyperparameter tuning. The optimization process focuses on key training parameters, including learning rate, batch size, and dropout rate, to improve convergence stability and reduce the risk of suboptimal local minima. The framework was evaluated on a balanced four-class waste image dataset (paper, wood, food, and leaves; N = 4000) using a five-fold cross-validation protocol. Experimental results achieved an average accuracy of 95.5% (±0.007), a macro-averaged AUC-ROC of 0.975, and a Cohen’s Kappa coefficient of 0.938, indicating strong agreement between predicted and true labels. Comparative experiments against ResNet-50 and a baseline ViT configuration suggest that SI-SSO optimization improves performance stability with only a modest increase in computational cost. These findings highlight the potential of optimized Transformer-based approaches for automated waste image classification under controlled evaluation conditions. Full article

Accurate and timely mapping of paddy rice is essential for agricultural management, food security, and climate-resilient policy. However, high-precision mapping remains challenging in subtropical monsoon regions due to persistent cloud cover, long revisit intervals, and striping noise, which compromise satellite data quality and availability. To address these limitations, a rice mapping framework suitable for different geographical environments was developed based on a random forest (RF) by combining time-series harmonic analysis (HANTS) with Sentinel-1 and Sentinel-2 multi-source data. To address these limitations, a rice mapping classification algorithm for different geographical environments was developed by combining Harmonic Analysis of Time Series (HANTS) with Sentinel-1/2 multi-source data. The research obtained annual maps of single-season and double-season rice in the research area from 2019 to 2024, with a spatial resolution of 10 m. The results indicated that the Sentinel-1, Sentinel-2, GEE, and HANTS algorithm can effectively support the yearly mapping of single- and double-season paddy rice in Anhui Province, China. The resultant paddy rice map has a high accuracy with overall accuracies exceeding 92% and Kappa coefficients above 0.84. HANTS effectively captures key phenological features of paddy rice, and it can especially enhance the discrimination between single- and double-season rice; compared to existing rice mapping products, the proposed approach reduces classification errors by an average of 3.92% in six major rice-producing cities, each with cultivation areas exceeding 1 million hectares; spatial correlation analysis indicates substantial heterogeneity in rice cultivation patterns across northern, central, and southern Anhui, associated with both biophysical and anthropogenic factors. These results indicate that integrating phenological data with machine learning can enhance the accuracy of long-term, high-resolution crop monitoring, and annual rice maps will offer valuable support for food security assessment, water resource management, and policy planning. Full article

Online signature verification (OSV) is a challenging problem in behavioral biometrics, especially when skilled forgeries closely mimic genuine signatures in both appearance and dynamics. This study presents a multi-class OSV framework that combines hybrid statistical features and nonlinear dimensionality reduction using Uniform Manifold Approximation and Projection (UMAP). A 40-dimensional feature set is created from statistical moments of dynamic writing parameters in both time and frequency (DCT) domains. Experimental results show that UMAP-based dimensionality reduction preserves category-related structures in a compact two-dimensional space. The proposed approach achieves an average classification accuracy of 0.989 ± 0.005 and a Cohen’s kappa coefficient of 0.985 ± 0.006, demonstrating robust performance across multiple classifiers. The results validate the effectiveness of combining multi-domain statistical feature fusion with UMAP for multi-class online signature verification, providing both high performance and interpretable visual representations. Full article

We envision a minimalist way to explain a number of astronomical facts associated with the unsolved missing mass problem by considering a new phenomenological paradigm. In this model, no new exotic particles need to be added, and the gravity is not modified; it is the perception that we have of a purely Newtonian (or purely Einsteinian) Universe, dubbed the Newton basis or Einstein basis (actually “viewed through a pinhole” which is “optically” distorted in some manner by a so-called magnifying effect). The $\kappa$ model is not a theory but rather an exploratory technique that assumes that the sizes of the astronomical objects (galaxies and galaxy clusters or fluctuations in the CMB) are not commensurable with respect to our usual standard measurement. To address this problem, we propose a rescaling of the lengths when these are larger than some critical values, say >100 pc - 1 kpc for the galaxies and ∼1 Mpc for the galaxy clusters. At the scale of the solar system or of a binary star system, the $\kappa$ effect is not suspected, and the undistorted Newtonian metric fully prevails. A key point of an ontological nature rising from the $\kappa$ model is the distinction which is made between the distances depending on how they are obtained: (1) distances deduced from luminosity measurements (i.e., the real distances as potentially measured in the Newton basis, which are currently used in the standard cosmological model) and (2) even though it is not technically possible to deduce them, the distances which would be deduced by trigonometry. Those “trigonometric” distances are, in our model, altered by the kappa effect, except in the solar environment where they are obviously accurate. In outer galaxies, the determination of distances (by parallax measurement) cannot be carried out, and it is difficult to validate or falsify the kappa model with this method. On the other hand, it is not the same within the Milky Way, for which we have valuable trigonometric data (from the Gaia satellite). Interestingly, it turns out that for this particular object, there is strong tension between the results of different works regarding the rotation curve of the galaxy. At the present time, when the dark matter concept seems to be more and more illusive, it is important to explore new ideas, even the seemingly incredibly odd ones, with an open mind. The approach taken here is, however, different from that adopted in previous papers. The analysis is first carried out in a space called the Newton basis with pure Newtonian gravity (the gravity is not modified) and in the absence of dark matter-type exotic particles. Then, the results (velocity fields) are transported into the leaves of a bundle (observer space) using a universal transformation associated with the average mass density expressed in the Newton basis. This approach will make it much easier to deal with situations where matter is not distributed centrosymmetrically around a center of maximum density. As examples, we can cite the interaction of two galaxies or the case of the collision between two galaxy clusters in the bullet cluster. These few examples are difficult to treat directly in the bundle, especially since we would include time-based monitoring (with an evolving $\kappa$ effect in the bundle). We will return to these questions later, as well as the concept of average mass density at a point. The relationship between this density and the coefficient $\kappa$ must also be precisely defined. Full article

Background: The original Closed Kinetic Chain Upper Extremity Stability Test (CKCUEST) is a simple assessment tool but does not account for individual differences in hand starting position and fails to provide information on limb asymmetries. The purpose of the study is to evaluate the test–retest reliability of a new single-arm CKCUEST as well as the reliability of the limb symmetry index (LSI). This version normalizes the test based on the participant’s arm length and allows for the assessment of limb symmetry since it is performed one arm at a time. Methods: Twelve healthy young adults provided both verbal and written consent to participate. Participants were excluded if they had sustained an injury in the past three months requiring medical attention and/or resulting in decreased activity for more than three days. Testing was conducted in the push-up position with participants’ thumbs placed parallel and at a distance equal to the length of their dominant arm (measured from the acromion to the tip of the middle finger), and feet positioned shoulder-width apart. Participants were instructed to keep the testing hand stable on the floor while the opposite hand reached across the body to touch the stationary hand and then return to the starting position marked with athletic tape. The goal was to complete as many touches as possible in 15 s, with each touch counted only if the participant touched the stationary hand, returned to the starting position, and maintained the shoulder-width stance. The average number of touches from the three trials was used for analysis. Intraclass Correlation Coefficients (ICC_(3,1)) were computed to determine test–retest reliability. Results: Test–retest reliability of the single-arm CKCUEST individual tests was good to excellent. The ICC_(3,1) was 0.88 (95% CI: 0.74–0.95) for all tests, 0.89 (95% CI: 0.66–0.96) for the dominant arm, and 0.93 (95% CI: 0.78–0.98) for the non-dominant arm. In contrast, the reliability of the Limb Symmetry Index (LSI) was questionable, showing substantial variability with an ICC_(3,1) of 0.53 (95% CI: −0.03–0.83) between Day 1 and Day 2, despite similar mean values (Day 1: 93.6 ± 8.46; Day 2: 94.8 ± 5.77). The Kappa coefficient suggested a substantial level of agreement for the direction of the asymmetry (preferred limb) (Kappa coefficient = 0.62). Conclusions: The new single-arm CKCUEST, which personalizes the hand starting position and measures limb symmetry, demonstrates high reliability among healthy young adults. Full article

Background/Objectives: Arboviral diseases share common vectors, geographic distribution, and symptoms. Developing Machine Learning diagnostic tools for co-circulating arboviral diseases faces data-scarcity challenges. This study aimed to demonstrate that proof of concept using synthetic data can establish computational feasibility and guide future real-world validation efforts. Methods: We assembled a synthetic dataset of 28,000 records, with 7000 for each disease—Dengue, Zika, and Chikungunya—plus Influenza as a negative control. These records were obtained from the existing literature. A binary matrix with 67 symptoms was created for detailed statistical analysis using Odds Ratios, Chi-Square, and symptom-specific conditional prevalence to validate the clinical relevance of the simulated data. This dataset was used to train and evaluate various algorithms, including Multi-Layer Perceptron (MLP), Narrow Neural Network (NN), Quadratic Support Vector Machine (QSVM), and Bagged Tree (BT), employing multiple performance metrics: accuracy, precision, sensitivity, specificity, F1-score, AUC-ROC, and Cohen’s kappa coefficient. Results: The dataset aligns with the PAHO guidelines. Similar findings are observed in other arboviral databases, confirming the validity of the synthetic dataset. A notable performance across all evaluated metrics was observed. The NN model achieved an overall accuracy of 0.92 and an AUC above 0.98, with precision, sensitivity, and specificity values exceeding 0.85, and an average Uniform Cohen’s Kappa of 0.89, highlighting its ability to reliably distinguish between Dengue and Influenza, with a slight decrease between Zika and Chikungunya. Conclusions: These models could accelerate early diagnosis of arboviral diseases by leveraging encoded symptom features for Machine Learning and Deep Learning approaches, serving as a support tool in regions with limited healthcare access without replacing clinical medical expertise. Full article

Because Internet of Things (IoT) networks are widely deployed, they have become attractive targets for botnet and distributed denial of service (DDoS) attacks, which require effective intrusion detection. Convolutional neural networks (CNNs) can achieve strong detection performance, but their many hyperparameters are usually tuned manually, which is costly and time-consuming. This paper proposes a new hybrid metaheuristic optimizer, FW-CNN, that combines Grey Wolf Optimization and Red Fox Optimization to automatically tune the key hyperparameters of a one-dimensional CNN for IoT intrusion detection. The Red Fox component enhances exploration and helps the search escape local optima, while the Grey Wolf component strengthens exploitation and guides convergence toward high-quality solutions. The proposed model is evaluated using the N-BaIoT dataset and compared with a feedforward neural network as well as a metaheuristic-optimized model based on the Adaptive Particle Swarm Optimization–Whale Optimization Algorithm-CNN. It achieves a final accuracy of 95.56%, improving on the feedforward network by 12.56 percentage points and outperforming the Adaptive Particle Swarm Optimization–Whale Optimization Algorithm-based CNN model by 1.02 percentage points. It also yields higher average precision, Kappa coefficient, and Jaccard similarity, and significantly reduces Hamming loss. These results indicate that the proposed hybrid optimizer is stable and effective for multi-class IoT intrusion detection in real environments. Full article

Terraces are the main engineering of soil erosion control on the Loess Plateau, offering measures for sediment reduction and water conservation, as well as the potential for increasing agricultural productivity. Over the years, large-scale terrace construction has been undertaken; however, the management has been inadequate, especially in terms of long-term monitoring and mapping. Moreover, the sediment reduction effect of terrace construction is not yet fully understood. Therefore, this study utilizes Landsat series data, integrating remote sensing imaging principles with machine learning techniques to achieve long–term temporal sequence mapping of terraces at a 30 m spatial resolution on the Loess Plateau. The sediment reduction effect brought about by terrace construction on the Loess Plateau is quantified using a sediment reduction formula. The results show that Elevation (Ele.), red band (R), Normalized Difference Vegetation Index (NDVI), Enhanced Vegetation Index (EVI), and Near-infrared Reflectance of Vegetation (NIRv) are key parameters for remote sensing identification of terraces. These five remote sensing variables explain 88% of the terrace recognition variance. Coupling the Random Forest classification model with the LandTrendr algorithm allows for rapid time-series mapping of terrace spatial distribution characteristics on the Loess Plateau. The producer’s accuracy of terrace identification is 93.49%, the user’s accuracy is 93.81%, the overall accuracy is 88.61%, and the Kappa coefficient is 0.87. The LandTrendr algorithm effectively removes terraces affected by human activities. Terraces are mainly distributed in the southeastern Loess areas, including provinces such as Gansu, Shaanxi, and Ningxia. Over the past 30 years, the terrace area on the Loess Plateau has increased from 0.9790 million hectares in 1990 to 9.8981 million hectares in 2020. The sediment reduction effect is particularly notable, with an average reduction of 49.75% in soil erosion across the region. This indicates that terraces are a key measure for soil erosion control in the region and a critical strategy for improving farmland productivity. The data from this study provides scientific evidence for soil erosion control on the Loess Plateau and enhances the precision of terrace management. Full article

The increasing rate of land use change, particularly deforestation and agricultural expansion, has intensified soil degradation, leading to reduced sediment retention and accelerated soil erosion. This study aims to analyze soil erosion and sediment retention in the Lam Phra Phloeng (LPP) watershed, Thailand, using a coupled modelling approach integrating the Revised Universal Soil Loss Equation (RUSLE) and the Sediment Delivery Ratio (SDR) model from the Integrated Valuation of Ecosystem Services and Tradeoffs (InVEST) suite. Six land use classes (forest, cropland, rangeland, flooded vegetation, built-up areas, and water bodies) were identified using Sentinel-2 MSI satellite data, with a Random Forest (RF) classification algorithm achieving an overall accuracy of 91.3% (Kappa coefficient = 0.89). The results indicate that forested areas exhibit the highest sediment retention, whereas croplands and rangelands experience the most significant soil loss due to erosion. The RUSLE model estimated an average annual soil loss ranging between 50 and 90 tons/ha/year, with the highest erosion rates observed in agricultural lands with steep slopes and minimal vegetation cover. The InVEST SDR model further corroborates these findings, showing that sediment retention is predominantly concentrated in densely vegetated areas, reinforcing the crucial role of natural forests in preventing soil displacement. This complementary modelling approach identifies priority areas for soil conservation practices. This study is the first study to integrate the RUSLE and InVEST models for the Lam Phra Phloeng watershed, providing a coupled assessment of erosion risk and sediment retention capacity and offering a novel and transferable framework for watershed-scale conservation planning and soil management in tropical monsoonal environments. Full article

Hyperspectral imagery (HSI), characterized by the integration of both spatial and spectral information, is widely employed in various fields, such as environmental monitoring, geological exploration, precision agriculture, and medical imaging. Hyperspectral image classification (HSIC), as a key research direction, aims to establish a mapping relationship between pixels and land-cover categories. Nevertheless, several challenges persist, including difficulties in feature extraction, the trade-off between effective integration of local and global features, and spectral redundancy. We propose FreqMamba, a novel model that efficiently combines CNN, a custom attention mechanism, and the Mamba architecture. The proposed framework comprises three key components: (1) A novel multi-scale deformable convolution feature extraction module equipped with spectral attention, which processes spectral and spatial information through a dual-branch structure to enhance feature representation for irregular terrain contours; (2) a novel group-separated attention module that integrates group convolution with group-separated self-attention, effectively balancing local feature extraction and global contextual modeling; (3) a newly introduced bidirectional scanning Mamba branch that efficiently captures long-range dependencies with linear computational complexity. The proposed method achieves optimal performance on multiple benchmark datasets, including QUH-Tangdaowan, QUH-Qingyun, and QUH-Pingan, with the highest overall accuracy reaching 97.47%, average accuracy reaching 93.52%, and a Kappa coefficient of 96.22%. It significantly outperforms existing CNN, Transformer, and SSM-based methods, demonstrating its effectiveness, robustness, and superior generalization capability. Full article

In the São Brás de Alportel municipality, water scarcity poses a significant constraint on agricultural activities. This study utilises Remote Sensing (RS) and Geographical Information Systems (GISs) to identify existing irrigated areas, delineate catchment basins, and select the most suitable sites for the installation of new surface water reservoirs. First, the principal territorial components were characterised, including physical elements (climate, geology, soils, and hydrography) and anthropogenic infrastructure (road network and high-voltage power lines). Summer Sentinel-2 satellite imagery was then analysed to calculate the Normalised Difference Vegetation Index (NDVI), enabling the identification and classification of irrigated agricultural parcels. Flow directions and accumulations derived from Digital Elevation Models (DEMs) facilitated the characterisation of 38 micro-catchments and the extraction of 758 km of the drainage network. The siting criteria required a minimum setback of 100 m from roads and high-voltage lines, excluded farmland currently in use, and favoured mountainous areas with low permeability. Only 18.65% (2854 ha) of the municipality is agricultural land, of which just 4% (112 ha) currently benefits from irrigation. The NDVI-based classification achieved a Kappa coefficient of 0.88, indicating high reliability. Three sites demonstrated adequate storage capacity, with embankments measuring 8 m, 10 m, and 12 m in height. At one of these sites, two reservoirs arranged in a cascade were selected as an alternative to a single structure exceeding 12 m in height, thereby reducing environmental and landscape impact. The reservoirs fill between October and November in an average rainfall year and between October and January in a dry year, maintaining a positive annual water balance and allowing downstream plots to be irrigated by gravity. The methodology proved to be objective, replicable, and essential for the sustainable expansion of irrigation within the municipality. Full article