Search Results (2,191)

Operational landslide susceptibility mapping (LSM) remains challenging in regions with pronounced geo-environmental heterogeneity, where single global models often overlook spatially variable landslide-environment relationships. Northern Guangdong, China, is a typical humid mountainous region where steep terrain, diverse lithology, and highly variable rainfall produce non-stationary landslide controls. To address this challenge, we develop a cluster-informed LSM framework that integrates unsupervised consensus K-means sub-zoning with localized Random Forest (RF) models and SHapley Additive exPlanations (SHAP). We use a harmonized inventory of 1510 landslides (2011–2022), together with twelve 30 m conditioning factors, for model training and validation. Compared with logistic regression, Support Vector Machines (SVM), and Light Gradient Boosting Machine (LightGBM), RF consistently achieves higher accuracy across clusters, and the cluster-wise RF ensemble attains pooled ACC = 0.8212, F1 = 0.8176, and AUC = 0.8956. SHAP highlights both regionally consistent predictors (e.g., NDVI, distance to road) and distinct cluster-specific controls linked to geomorphic and hydrologic settings. The proposed framework enhances predictive accuracy, produces finer susceptibility gradients, and yields better-calibrated probability estimates than a single global model. These results demonstrate that explicitly accounting for geo-environmental heterogeneity can generate interpretable, spatially adaptive susceptibility outputs. By identifying high-risk zones for priority monitoring, land-use regulation, infrastructure protection, and mitigation planning, the proposed framework provides a practical decision-support tool for sustainable mountain development and disaster risk reduction in heterogeneous mountainous regions. Full article

In low-light, occluded, and cluttered environments, single-modality RGB detectors are prone to false positives and missed detections. While infrared (IR) imaging provides relatively stable target visibility under poor illumination, it lacks texture and color information and is susceptible to background thermal noise and imaging variations. To address these limitations, this paper proposes an RGB–IR object detection network, named MCF-YOLO, consisting of three core components. First, the Cross-Modal Hierarchical Fusion (CMHF) module performs stage-wise alignment and fusion on multi-scale features, jointly modeling RGB texture details and IR thermal responses to exploit the structural and semantic complementarity between the two modalities. Second, the Soft Attention Regularization based on Attention Prior (SAR-AP) module derives attention priors from IR features to impose soft constraints on cross-modal attention maps. This mechanism helps the network maintain attention on target-relevant regions, thereby suppressing attention drift caused by low-light noise and complex backgrounds. Third, the Small-Object-Sensitive Detection Head (SOS-Head) processes high-resolution features to strengthen the representation of small targets, improving detection capability in long-range and occluded scenarios. In evaluations on two RGB–IR benchmarks—M3FD and VEDAI—MCF-YOLO achieves improvements of 2.7% in mAP@0.5 and 1.1% in mAP@0.5:0.95 on M3FD, and 5.4% and 4.4%, respectively, on VEDAI. These results suggest that consistency-guided cross-modal fusion and high-resolution small-target modeling are beneficial for RGB–IR detection in low-visibility and cluttered scenes. Full article

Urban flood vulnerability is increasingly shaped by the interaction between climate change, urbanization, and spatial planning practices, particularly in Mediterranean metropolitan areas. This study develops an integrated GIS–AHP framework to assess the susceptibility component of flood vulnerability in the urban area of Thessaloniki, Greece. Using open-access geospatial data, ten indicators representing soil, hydrological, and environmental conditions are derived and spatially analyzed. The Analytic Hierarchy Process (AHP), based on expert judgment, is applied to estimate the relative importance of these indicators and to support their integration into a composite flood susceptibility index. The results reveal strong spatial heterogeneity, with high susceptibility concentrated in low-lying, densely urbanized areas and zones near drainage pathways. Among the examined factors, the Topographic Wetness Index emerges as the most influential, highlighting the persistent role of terrain-driven hydrological processes even in highly built environments. The proposed framework provides a transparent and transferable methodology for identifying flood-prone areas and supports evidence-based urban planning and climate resilience strategies. The findings contribute to the broader discussion on vulnerability and resilience in urban systems by linking spatial analysis with decision-support tools in a policy-relevant context. Full article

The Gaizi River Basin, an alpine region in China crossed by the Karakoram Highway, is highly prone to glacier-related debris flows (GDF). Accurate debris flow susceptibility assessment in this high-altitude area remains challenging due to complex terrain, active tectonics, and dynamic glacial processes. This study develops a hybrid model integrating statistical methods and machine learning-based anomaly detection for debris flow susceptibility mapping. To address data noise, certainty factor (CF) distributions of debris flow predisposing factors (DFPFs) were derived via Locally Weighted Scatterplot Smoothing (LOWESS). The strength of the association between DFPFs and GDF susceptibility was evaluated using the mean residual between the raw and LOWESS-smoothed CF values. Multiple anomaly detection algorithms, including distance-based (L2 Norm), density-based (One-Class SVM), ensemble (Isolation Forest, RandNet), and GAN-based (WBiGAN-GP) methods, were tested on raw and CF-transformed data, using only the GDF inventory as the label. The CF-WBiGAN-GP model delivers the most balanced performance, excelling at identifying both high- and low-susceptibility zones. Results show that distance to stream, slope, and the topographic roughness and wetness indices are strongly associated with GDF susceptibility. Distance to glacier and precipitation appear less informative for direct susceptibility inference under our specific dataset and analytical setup. Full article

The increasing number of extreme events affecting buildings and strategic infrastructures in mountain areas requires reliable approaches for territorial risk assessment with respect to snow avalanches. Considering risk as the combination of hazard, vulnerability, and exposure factors, a simplified framework—recently adopted in Italian national guidelines—is proposed. Avalanche hazard is defined by considering the intrinsic avalanche susceptibility of the territory under investigation, typically described by means of hazard intensity maps. On the other hand, the vulnerability of the construction is determined by considering both the physical, or structural, vulnerability of the building and the functional vulnerability of network systems. Finally, the exposure level accounts for the direct and indirect losses resulting from the hazardous event, based on the typology, use, and potential occupancy of the building or infrastructure. A weighted classification system combining these three factors is adopted to derive risk matrices, in which the risk class of each exposed construction is defined across five levels (high, medium–high, medium, medium–low, low), thus enabling a hierarchical risk classification at the territory scale. The methodology is intended to bridge technical risk assessment and territorial governance, offering an operational decision-support tool for policymakers, emergency planners, and infrastructure operators to support resource allocation and mitigation strategies. Full article

Landslides constitute a globally pervasive and highly destructive natural hazard. Although artificial intelligence (AI)-driven landslide susceptibility mapping has emerged as an effective tool for delineating high-risk zones, its predictive performance is frequently constrained by inherent data noise and insufficient characterization of landslide triggering factors, restricting the credibility of the mapping results. In this study, to remedy this limitation, we adopt fractal analysis to extract latent inherent information from topographic features. Specifically, the box-counting method and multifractal analysis are applied to excavate the intrinsic nonlinear characteristics embedded in eight topographic factors, and an improved K-means algorithm is utilized to perform feature selection and construct a dedicated fractal feature dataset, which is fed to advanced AI models. Our results indicate that the information dimension ($D_1$) of the slope gradient, the correlation dimension ($D_2$) of aspect, land relief, the $D_2$ of roughness, the $D_2$ of plan curvature, the multifractal spectrum width ($∆\mathsf{\alpha }$) of profile curvature, the $D_2$ of elevation, and the surface cutting depth were the most effective features, demonstrating superior performance in capturing landslide targets. Comparative performance evaluations reveal that AI models trained on fractal features demonstrate substantially superior predictive capabilities compared to AI models trained on raw features. This superiority is consistently evidenced across key evaluation metrics, including overall accuracy, kappa coefficient, F1-score, and predictive efficiency, demonstrating that the integration of fractal characteristics significantly augments model robustness and predictive efficacy. To mitigate the ‘black-box’ problem of AI modeling, Shapley additive explanations were employed to quantify individual feature contributions and elucidate the underlying predictive mechanisms. Our findings indicate that the integration of fractal analysis yields highly discriminative and robust feature representations, thereby expanding the representational capacity of the models and improving predictive accuracy. Furthermore, a joint assessment of spatial uncertainty and susceptibility maps demonstrates that these models exhibit low predictive variance and high spatial stability when delineating high-susceptibility zones. Notably, models utilizing fractal-derived features achieve superior spatial capture efficiency. The resultant topographic features characterized by fractal representation and selected via the improved K-means algorithm can significantly improve the predictive performance of trained AI models in landslide susceptibility mapping tasks, offering a scientific and viable technical approach for future landslide prediction and prevention. Full article

Liquefaction phenomena are strongly influenced by the depositional evolution of the area, including sediment grain size, depositional age, shallow layering, and groundwater depth. This study focuses on a 560 km² wide sector of the eastern Po River Plain (northern Italy), encompassing part of the modern Po Delta, to evaluate the susceptibility of the different geological units to liquefaction. A comprehensive dataset was compiled, integrating lithological, chronological (¹⁴C), geomorphological, hydrological, and hydrogeological information, together with satellite imagery, historical and modern maps, archaeological evidence, and subsurface data from core drilling and CPTu tests. The integrated analysis allowed us to reconstruct a liquefaction susceptibility map recognizing four classes: very high (4% of the investigated area), high (26%), moderate (20%), and non-susceptible (50%). CPTu-based statistical analyses confirm that the Liquefaction Potential Index (LPI) increases with higher susceptibility classes and decreases with increasing groundwater depth (0.5, 1.5, and 3.0 m scenarios). These results provide a scientific basis to support sustainable land management and governance strategies in the Po Delta, an area of high environmental, cultural, and economic value, a large sector of which is included in the Natura 2000 network. Full article

Background/Objectives: Banana (Musa spp.) production is severely threatened by Fusarium wilt caused by Fusarium oxysporum f. sp. cubense tropical race 4 (Foc TR4), highlighting the need to identify genetic determinants of resistance. Methods: We performed a genome-wide analysis of NBS genes in Musa acuminata ssp. malaccensis, including phylogenetic, chromosomal, and microsynteny analyses. The genomic context and promoter regions of MamRGA2 were characterized, its response to defense-related phytohormones was evaluated by RT-qPCR, and its protein structure was predicted by homology modeling. Results: A total of 118 NBS genes were identified. Notably, we report for the first time in banana two NBS genes encoding proteins with integrated domains, corresponding to an ATP-binding cassette (ABC) transporter and a Nuclear Factor Y subunit A (NF-YA) transcription factor. Chromosomal mapping revealed a marked enrichment of NBS genes on chromosome 3, which harbors MamRGA2, an NBS gene associated with resistance to Foc TR4. RT-qPCR analyses showed that MamRGA2 is strongly induced by exogenous methyl jasmonate (MeJA) in the resistant wild genotype but not in a susceptible Cavendish cultivar, a pattern associated with divergence in promoter sequences between the two genotypes. Structural modeling suggested that the MamRGA2 protein possesses features consistent with a resistosome-like architecture. Conclusions: Overall, these findings expand current knowledge of NBS gene diversity in banana and provide a framework for future studies aimed at elucidating the molecular mechanisms underlying resistance to Foc TR4. Full article

Landslides cause thousands of fatalities and billions in economic losses annually, yet reliable susceptibility mapping across heterogeneous landscapes remains challenging because conventional models assume stationary relationships between landslide occurrence and environmental controls. Ensemble methods, though promising, rely on either globally fixed aggregation weights or kernel-constrained local averaging, failing to adapt when the reliability of base models varies nonlinearly across space. To overcome this, we propose a two-stage hierarchical spatial adaptive ensemble (HSE) framework. In stage one, three complementary base learners are deployed: geographically weighted regression (GWR) for local spatial non-stationarity; a geographically optimal similarity (GOS) model, grounded in the Third Law of Geography, to represent similarity-based local dependence; and a deep neural network (DNN) for nonlinear covariate interactions. In stage two, a multi-branch attention-based network learns spatially varying fusion weights via multi-scale feature extraction, abandoning fixed weights or kernel constraints. We validate HSE on a typical landslide-prone catchment, comparing against single models (GWR, DNN, GOS). Results demonstrate that our method consistently achieves superior predictive accuracy, spatial consistency, and out-of-sample robustness. Moreover, the attention-derived spatially varying weights provide interpretable insights into where each base learner dominates, bridging predictive performance with geophysical interpretability. These findings confirm that explicitly learning spatial heterogeneity during ensemble fusion is essential for reliable landslide susceptibility mapping, with strong potential for transfer to other geospatial prediction tasks. Full article

Fish phenotyping plays an important role in growth evaluation, selective breeding, and precision aquaculture. Conventional phenotypic measurement methods are labor-intensive, time-consuming, and susceptible to observer variability. To improve measurement efficiency and reproducibility, this study proposes an automated fish phenotyping framework based on Transformer-enhanced instance segmentation. Specifically, a Mask2Former decoder was integrated into the Mask R-CNN architecture to improve boundary delineation and segmentation quality. Based on segmentation outputs, phenotypic parameters, including body length, body height, and projected area, were automatically extracted using PCA-assisted orientation estimation and geometric measurement. In addition, a standardized anatomical landmark annotation framework consisting of 12 reference points was introduced to support reproducible phenotypic description and future extensible morphometric analysis. Body weight was further estimated using polynomial regression based on extracted morphological traits. Experiments were conducted using images from three crucian carp varieties under controlled imaging conditions. The proposed framework achieved 92.7% mAP and 89.4% Boundary IoU, improving segmentation performance over the baseline model. Automated measurement yielded average relative errors of 2.16% for body length and 3.85% for body height, while weight prediction achieved an R² of 0.9479 and a mean relative error of 7.31%. These results demonstrate that Transformer-enhanced segmentation can support accurate and efficient automated phenotyping under standardized conditions and provide a foundation for future deployment in more complex aquaculture environments. Full article

Host genetic susceptibility to urogenital Chlamydia trachomatis (Ct) reinfection remains poorly understood. Coding variants identified in prior genome-wide association studies (GWAS) explained only a small fraction of the risk of reinfection. Our goal in this study was to characterize whether more risk would be captured by sequence variation that traditional GWAS insufficiently captures. Specifically, we evaluated the risk attributable to SNPs present in regulatory, non-coding regions; post-transcriptional regulation by microRNAs (miRNAs) that may depend on sequence variation in either the miRNA or the target mRNA; and copy number variants (CNVs). We analyzed GWAS data from African American women with or without documented urogenital Ct reinfection. Fine mapping and independent association analyses identified 30 unique index single-nucleotide polymorphisms (iSNPs), which were expanded to variants in linkage disequilibrium. Regulatory annotation was performed using HaploReg, RegulomeDB, FORGEdb, rSNPBase, and GTEx. We examined whether genes identified in the Ct reinfection GWAS are targeted by known Ct infection–associated microRNAs using curated databases. Genome-wide CNV calling was conducted using SNP intensity data, followed by stringent quality control and gene-level association testing. Functional annotation prioritized 7 SNPs with strong regulatory evidence, with stringent criteria for regulatory relevance, using HaploReg, RegulomeDB, FORGEdb, and rSNPBase. The strongest signals were observed at the CHIT1 locus, where multiple intronic variants (including rs2486963 and rs2244385) overlapped regulatory chromatin, altered transcription factor binding motifs, and acted as cis-expression quantitative trait loci for CHIT1 in whole blood. Additional regulatory variants were identified near TDRP, ERICH1, and DLGAP1, showing tissue-specific regulatory effects. MicroRNA analysis revealed extensive post-transcriptional targeting of SOCS6 and SULF1, while CHIT1 showed no curated Ct-associated miRNA interactions. CNV analysis identified 5775 high-confidence events, with nominal gene-level associations observed for ATAD3A, CARD14, TMEM240, and ZNF140. These results indicate that a greater fraction of the susceptibility to urogenital Ct reinfection may be driven by genetic variation affecting immune and epithelial pathways rather than protein-coding changes. Full article

In this study, a soil erosion intensity map for the territory of Serbia was produced using the Modernized Erosion Intensity Coefficient (MEIC-Z), combined with remote sensing data (Sentinel-2) and Geographic Information Systems (GIS). The analysis was based on contemporary geospatial data on lithology, land use, and terrain slope, with a spatial resolution of 30 m. Particular emphasis was placed on modifying the φ coefficient, which significantly improved estimates of erosion intensity. The average erosion intensity at the national level is 0.239, corresponding to the weak erosion class. Multivariate analysis of geographical conditions showed that the highest values of the erosion coefficient (Z) were determined by agricultural land (r = 0.826), while the lowest values were associated with terrain slope (r = −0.805) and forest cover (r = −0.767). In addition to the national-scale assessment, spatial differentiation of the results was performed at the local (municipal) level. Municipalities were differentiated into four clusters using Agglomerative Hierarchical Clustering. The advantage of the modified φ coefficient lies in the integration of land use and terrain slope, enabling a more realistic assessment of the intensity of erosion processes. Validation results demonstrated strong agreement between the modernized Z-derived erosion coefficient and the expert-defined erosion inventory, supporting the internal consistency of the model-derived erosion susceptibility patterns. This study significantly contributes to decision-making at both national and local levels by providing a scientific basis for developing strategies for sustainable forest management and soil conservation. Full article

The Greater Khingan, a key cold-temperate coniferous forest region in northern China, is frequently affected by forest fires with severe ecological and economic impacts. The study investigates the influence of key environmental and anthropogenic drivers on forest fire susceptibility and evaluates multiple machine-learning approaches for regional fire assessment. Using 2001–2018 fire point data and multi-source remote sensing data, we integrated 13 driving factors across four dimensions: meteorology, topography, vegetation, and human activities. Collinear variables were screened using the Variance Inflation Factor (VIF). Three machine learning models—Logistic Regression (LR), Random Forest (RF), and Support Vector Machine (SVM)—were constructed to assess the long-term potential risk of forest fire occurrence. Driving mechanisms were analyzed using standardized regression coefficients and the SHapley Additive exPlanations (SHAP) interpretable algorithm, and spatial distribution maps of regional forest fire risk were generated based on the optimal model. Among the three models, RF achieved the highest predictive accuracy, with an accuracy of 0.919 and an Area Under the Receiver Operating Characteristic Curve (AUC) of 0.966, significantly outperforming LR and SVM. SHAP analysis reveals that forest fires are primarily driven by climatic factors (Pres and Prec as core drivers), regulated by topographic factors, and weakly affected by human factors. The proposed framework provides an effective tool for long-term forest fire susceptibility assessment by combining robust predictive performance with interpretable model outputs. The findings provide scientific support for long-term strategic forest fire risk zoning, regional firefighting resource allocation, and the formulation of differentiated prevention and control strategies, and also offer methodological references for forest fire prediction in other cold-temperate forest regions in China. Full article

Landslide susceptibility algorithms demonstrate high reliability in quantifying the likelihood of landslide occurrence. However, traditional methods are often limited by computationally intensive sampling strategies and models with limited adaptability. In this study, we propose an adaptive sampling strategy based on hotspot analysis to enhance the reliability of the generated samples. Additionally, we develop an improved meta-ensemble (IME) stacking-based heterogeneous framework for landslide susceptibility assessment by integrating a support vector machine (SVM), random forest (RF), and XGBoost. To further reduce factor complexity, a Monte Carlo-based frequency ratio analysis is employed. The Baihetan Reservoir area along the Jinsha River was selected as the study area. A total of 26 conditioning factors were considered, supplemented by 120 Sentinel-1A images to cover the study area. The proposed sampling strategy was then used to generate high-quality samples. Finally, to evaluate the performance of the proposed method, the proposed ensemble learning framework was applied to assess landslide susceptibility with eight models using five evaluation metrics. The experimental results demonstrated that: (1) the adaptive sampling strategy improved both the quantity and quality of the training samples; (2) the adoption of the Monte Carlo strategy increased the sample partitioning rate; and (3) despite the formally highest IME metrics, the inclusion of InSAR information did not lead to a statistically significant improvement in the forecast compared to the high-quality basic sampling strategy. Overall, the proposed methodology provides valuable support for regional geohazard susceptibility assessment in dynamic environments. Full article

The growing need for secure image transmission across public networks requires robust encryption algorithms. Traditional chaos-based image ciphers typically have a small key space, weak avalanche behavior, or are susceptible to differential cryptanalysis. To overcome such inadequacies, this paper suggests a new adaptive image cryptosystem that combines a fractional-order memristive chaotic engine and a non-linear hybrid encryption kernel. The system uses piano-inspired feedback; the keystream generator dynamically adapts to the previously encrypted pixel, enabling powerful Cipher Block Chaining (CBC)-style chaining and content-dependent diffusion. A four-dimensional memristive system is solved by the use of fractional-order calculus, which gives an ultra-large key space (>10⁸⁰) and very high sensitivity to initial conditions—confirmed by a positive largest Lyapunov exponent (1.7199). The encryption kernel maps the traditional Exclusive OR (XOR) with the reversible two-step operation: the modular addition of the plaintext with the first keystream byte and the XOR with the second keystream one, both of which increase non-linearity and confusion. Large-scale experiments with six standard 256 × 256 colour images indicate almost ideal entropy (7.9994), Number of Pixel Change Rate (NPCR) which is 99.62, Unified Average Changing Intensity (UACI) which is 33.43, correlation coefficients are near to zero, very low Gray-Level Co-occurrence Matrix (GLCM) homogeneity (≈0.017) and high contrast (≈4843) and low energy (≈0.006 The ciphertext passes seven National Institute of Standards and Technology (NIST) SP-800-22 statistical tests, is extremely sensitive to keys (a perturbation of 1 × 10⁻¹⁴ alters >99.6% of ciphertext) and resists chosen-plaintext and known-plaintext attacks. Decryption has linear time complexity O(N), and average encryption and decryption times are 3.40 s and 2.75 s for 256 × 256 images. The proposed cryptosystem provides an attractive security–performance trade-off that can be used in high-security systems like medical image protection, privacy-preserving multimedia transmission, and secure cloud storage. Full article