Egyptian Journal of Remote Sensing and Space Sciences最新文献

{"title":"SMF2Net: Spectral-assisted multi-receptive field fusion network for hyperspectral and multispectral image fusion","authors":"Siyuan Liu ,&nbsp;Huanru Yue ,&nbsp;Ruixia Cai ,&nbsp;Bing Li ,&nbsp;Yudong Zhang ,&nbsp;Shuaiqi Liu","doi":"10.1016/j.ejrs.2026.02.001","DOIUrl":"10.1016/j.ejrs.2026.02.001","url":null,"abstract":"<div><div>Hyperspectral image (HSI) is extensively used in classification, detection, tracking, and other tasks attributed to its capacity to capture extensive spectral information and comprehensively characterize the spectral signatures unique to distinct materials. Nevertheless, the acquisition of high-resolution HSI (HR-HSI) is fundamentally challenged by the intrinsic technical constraints of HSI sensors. Therefore, the fusion of low-resolution HSI and high-resolution multispectral image (MSI) to obtain HR-HSI has become a research hotspot. Existing fusion algorithms often do not make full use of the correlation between spatial and spectral information, which makes the fusion model lack interpretability. Therefore, this paper builds a spectral-assisted multi-receptive field fusion network (SMF2Net) for HSI and MSI fusion. Specifically, for spatial information, this paper designs a spectral feature-based spatial partition convolution block (SFSPB) and a multi-receptive field interaction fusion block (MFIFB) to capture the spatial information in the source images. Among them, the SFSPB extracts spatial information in different regions according to the spectral features of the image; meanwhile, the MFIFB extracts the spatial information of different receptive fields and performs feature fusion so that SMF2Net can obtain rich semantic information. For spectral information, this paper constructs a HybridFormer block based on spectral multi-head self-attention. It models the long-distance dependence of the spectrum through multi-head self-attention and enhances the spectral information in the fusion result through channel attention. In this paper, experiments are carried out on one real dataset and two simulated datasets. The experimental results indicate that the proposed algorithm can achieve advanced fusion effectiveness subjectively and objectively. The source code is available at: <span><span>https://github.com/cvmdsp/SMF2Net</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":48539,"journal":{"name":"Egyptian Journal of Remote Sensing and Space Sciences","volume":"29 1","pages":"Pages 147-157"},"PeriodicalIF":4.1,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146188081","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"AN MLPEL machine learning model for bathymetry retrieval based on ensemble learning","authors":"Jinshan Zhu ,&nbsp;Cong Jiao ,&nbsp;Ruifu Wang ,&nbsp;Yuquan Wen ,&nbsp;Yu Wang ,&nbsp;Bopeng Liu ,&nbsp;Yina Han","doi":"10.1016/j.ejrs.2026.01.003","DOIUrl":"10.1016/j.ejrs.2026.01.003","url":null,"abstract":"<div><div>Machine learning models have made rapid progress in recent years. As a feedforward neural network, the multilayer perceptron (MLP) is widely used in bathymetry because of its simple structure and good nonlinear fitting ability. However, the classical multilayer perceptron method is faced with a potential problem, which is easy to fall into local minima and overfitting, resulting in failure of model training. In order to solve the problems of overfitting and weak ability to capture complex environments of a single model, this paper propose a multilayer perceptron as the <em>meta</em>-model Ensemble Learning Model (MLPEL) In this model, CatBoost, MLP and Random Forest (RF) are selected for the base model to make preliminary prediction and generate new dataset. MLP model is selected as the <em>meta</em>-model to receive new dataset and predict water depth.<!--> <!-->To validate the feasibility of the model, we conducted experiments in two study areas, Ganquan Island and Vieques. In Ganquan island, the MSE of MLPEL model is 1.30 m², which is 0.6–2.4 m² lower than that of Log-Ratio (LR) model and MLP model. In Vieques island, the MSE of the MLPEL model is 1.54 m², which is 0.5–1.5 m² lower than the LR model and the MLP model. The results show that this method can solve the local minimum problem of multi-layer perceptron model well, and obtain the water depth value with high robustness and accuracy.</div></div>","PeriodicalId":48539,"journal":{"name":"Egyptian Journal of Remote Sensing and Space Sciences","volume":"29 1","pages":"Pages 15-27"},"PeriodicalIF":4.1,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145977640","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The Pacific Ocean decadal satellite data analysis and ENSO events connectivity to the Halmahera Sea","authors":"Muhammad Zainuddin Lubis ,&nbsp;Andrean V.H. Simanjuntak ,&nbsp;Batara ,&nbsp;Ekoué Ewane Blaise Arnold ,&nbsp;Song Hu","doi":"10.1016/j.ejrs.2026.01.006","DOIUrl":"10.1016/j.ejrs.2026.01.006","url":null,"abstract":"<div><div>The Pacific Ocean (PO) greatly influences the Indonesian sea, plays a crucial role in global climate and ocean circulation, and connects to the Indian Ocean (IO). Previous research has examined the Pacific Ocean’s influence, but the roles of eddies and primary productivity in marine ecosystems during the ENSO response within the Indonesian Throughflow (ITF) in the Halmahera Sea (HS) remain unclear. This study utilizes satellite data to analyze the spatiotemporal patterns of various oceanographic parameters with the ITF during ENSO events from 2009 to 2020. The analysis incorporates satellite-derived SST, SSH, SWH, Chl-a, zooplankton variability, and volume transport. Analysis of SST and SSH variability shows the influence of SWH. The EOF analysis reveals that SSH and SST variability are impacted by ENSO events, with notable fluctuations observed in 2010–2011 and 2015–2016. The EKE divergence is higher during La Niña than El Niño events. The volume transport is strongest at depths of 100–150 m during El Niño and 50–100 m during La Niña, with interannual variations ranging from −0.709 ± 1.394 Sv in the 0–50 m layer, decreasing with depth. Our findings show different patterns for Chl-a, SST, and SSS during the ENSO events.</div></div>","PeriodicalId":48539,"journal":{"name":"Egyptian Journal of Remote Sensing and Space Sciences","volume":"29 1","pages":"Pages 44-59"},"PeriodicalIF":4.1,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146037837","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Stacked ensemble model for flood layer extraction using EOS-04 satellite in Fine Resolution Stripmap (FRS) mode","authors":"Y.V. Sai Bhageerath,&nbsp;A.V. Suresh Babu,&nbsp;K.H.V. Durga Rao","doi":"10.1016/j.ejrs.2026.01.005","DOIUrl":"10.1016/j.ejrs.2026.01.005","url":null,"abstract":"<div><div>This study presents a novel methodology for enhancing flood detection in India using Fine Resolution Stripmap (FRS) mode imagery from the EOS-04 Synthetic Aperture Radar (SAR) satellite. The proposed approach integrates deep learning and classical machine learning techniques through a stacked ensemble framework designed for water body extraction. Five deep learning models are independently trained, and their outputs are combined to construct a new feature matrix, which is subsequently used as input for five traditional classifiers: Logistic Regression, XGBoost, Random Forest, K-Nearest Neighbors, and Support Vector Machine. Among these, the XGBoost classifier achieved the highest classification accuracy of 96.8 %. A Land-Use-Land-Cover (LULC) water mask, provided by the National Remote Sensing Centre (NRSC), was applied to delineate the flood layer. Model performance was validated through cross-comparison with flood maps derived from concurrent-date optical satellite imagery, yielding an overall accuracy of 96.5%. Furthermore, a pixel-wise uncertainty map was derived from the ensemble predictions to quantify model confidence, offering additional insight into spatial reliability of flood extent mapping. By leveraging the complementary strengths of deep and traditional learning approaches, this method addresses the inherent limitations of FRS-mode SAR data for flood detection. The proposed framework offers a robust and scalable solution for operational flood monitoring, with significant implications for disaster preparedness and response across flood-prone regions of India.</div></div>","PeriodicalId":48539,"journal":{"name":"Egyptian Journal of Remote Sensing and Space Sciences","volume":"29 1","pages":"Pages 60-71"},"PeriodicalIF":4.1,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146037838","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Air pollution mapping and monitoring using Sentinel-5P data and Google Earth Engine","authors":"Sara Sameh ,&nbsp;Ahmed Zaki ,&nbsp;Basem Elsaka ,&nbsp;Ashraf A.A. Beshr ,&nbsp;Ashraf G. Shehata","doi":"10.1016/j.ejrs.2026.01.008","DOIUrl":"10.1016/j.ejrs.2026.01.008","url":null,"abstract":"<div><div>Urbanization and industrialization have significantly altered land surface temperature (LST) and air quality (AQ), particularly in rapidly developing regions such as Sharkia Governorate, Egypt. This research presents a comprehensive spatio-temporal distribution of LST and air pollutants (APTs) interactions in Sharkia Governorate, Egypt, for 2024. LST data were derived from the MOD11A1 product, while PM2.5 and PM10 concentrations were obtained as AOD-derived estimates using empirical models calibrated in other regions, which are considered invalidated approximations pending local calibration. Sentinel-5P (S5P) data were used for APTs, and all datasets were processed through the Google Earth Engine (GEE). The southern industrial zones, particularly 10th of Ramadan and Belbies, exhibit significantly higher LST values, whereas northern rural areas maintain lower temperatures due to vegetation cover and agricultural activity. Correlation analysis reveals strong positive relationships between LST and key pollutants such as Methane (CH₄, r = 0.8589) and UVAI (r = 0.8874), indicating that higher temperatures enhance pollutant concentration and dispersion. A moderately positive correlation was found with Formaldehyde (HCHO, r = 0.4282). Conversely, Ozone (O₃) exhibited a moderately negative correlation (r = -0.7854) with LST. Particulate matter (PM2.5, PM10; r = 0.286, 0.413) showed weak correlations, suggesting that their emissions are primarily driven by anthropogenic sources rather than temperature fluctuations. Principal Component Analysis (PCA) was incorporated to explore multivariate relationships among pollutants and identify dominant underlying factors influencing air quality patterns. Geographically weighted regression (GWR) identified pollution hotspots in the southwest, particularly high-density urban and industrial areas, where CH₄ and UVAI showed stronger impacts on LST (local R² &gt; 0.8). All PM results are presented as invalidated estimates, and their interpretation is approached cautiously. The findings underscore the implications of urbanization and industrial activities on local climate, air quality, and public health, emphasizing the need for ground-based validation and mitigation policies.</div></div>","PeriodicalId":48539,"journal":{"name":"Egyptian Journal of Remote Sensing and Space Sciences","volume":"29 1","pages":"Pages 158-178"},"PeriodicalIF":4.1,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146188082","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Landslide Susceptibility Assessment Using an Explainable Stacking Learning Framework (ESLF) in Xinyuan County, Xinjiang, China","authors":"Yong Dai ,&nbsp;Shenglong Gu ,&nbsp;Qingkai Meng ,&nbsp;Shilong Chen ,&nbsp;Qiuhui Wang ,&nbsp;Ying Meng ,&nbsp;Han Wu ,&nbsp;Qing Li","doi":"10.1016/j.ejrs.2026.02.002","DOIUrl":"10.1016/j.ejrs.2026.02.002","url":null,"abstract":"<div><div>To address the lack of transparency in machine learning methods for landslide susceptibility assessment (LSA), this study proposes an Explainable Stacking Learning Framework (ESLF), taking Xinyuan County, Xinjiang, China as the study area. The framework effectively integrates Deep Neural Networks (DNN), Random Forests (RF), and Support Vector Machines (SVM), leveraging their respective strengths in pattern recognition, decision analysis, and hyperplane-based classification. A comprehensive landslide inventory and 11 predisposing factors were compiled to generate susceptibility maps through the application of Stacking, alongside individual DNN, RF, and SVM models. The results indicate that the stacking model outperforms single models, achieving AUC of 0.907, accuracy of 0.930, F1-score of 0.897, and a Kappa coefficient of 0.866. Very high and high susceptibility zones are mainly distributed in southern Talede Town, Biesituode Township, Xinyuan Town, Alemale Town, western Nalati Town, and northeastern parts of Zeketai, Areletuobie, Kansu Towns, and Tuergen Township. SHAP (SHapley Additive exPlanations) permutation importance analysis identifies elevation (995–2,253 m), distance to rivers (&lt;836 m), land use type (shrubs/woodlands or other types), and engineering geological lithology (clastic rocks) as dominant controlling factors. These findings highlight the ESLF’s advantage in improving both accuracy and transparency, providing civil protection agencies with a reliable tool for understanding landslide susceptibility and implementing effective mitigation measures.</div></div>","PeriodicalId":48539,"journal":{"name":"Egyptian Journal of Remote Sensing and Space Sciences","volume":"29 1","pages":"Pages 196-208"},"PeriodicalIF":4.1,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146188113","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Optimal feature-based InSAR phase filtering framework using convolutional neural network and mathematical morphology","authors":"Mahmoud S. Hamid ,&nbsp;Mohamed Ghetas ,&nbsp;Khaled M. Reda ,&nbsp;M. Safy","doi":"10.1016/j.ejrs.2026.01.001","DOIUrl":"10.1016/j.ejrs.2026.01.001","url":null,"abstract":"<div><div>Denoising interferometric phase data is a critical processing step for precise topographic reconstruction and land deformation analysis. A noisy interferogram leads to phase unwrapping challenges and may ultimately degrade the quality of various InSAR final products. In this paper, an optimal feature-based interferometric phase image filtering framework is proposed. The filtering process begins by extracting a set of features from simulated interferogram phase image that captures the primary structural elements of the interferometric phase data. Then, a set of soft morphological filters is optimized using the interferogram features set in the complex number field. A convolutional neural network is developed and employed to match patches of real interferogram phase images with the set of interferogram phase image features. The interferogram phase image patches are then filtered with their corresponding optimal filters. The convolutional neural network is trained using a dataset of interferogram phase image features with various noise variance. The proposed framework considers the effect of interferogram noise level on filtering performance by optimizing the filter parameters according to the level of the interferometric phase noise. Experimental results on simulated and real interferogram phase images confirm that the proposed interferogram filtering framework achieves efficient interferogram phase unwrapping as it contributed to high equivalent number of looks (ENL) of 5921 and a minimal number of phase residues (NOR) of 123. The results also approve that the proposed approach has perfect preservation of fringe pattern and interferogram fine detail compared to other state of the art approaches.</div></div>","PeriodicalId":48539,"journal":{"name":"Egyptian Journal of Remote Sensing and Space Sciences","volume":"29 1","pages":"Pages 28-43"},"PeriodicalIF":4.1,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145977564","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A geospatial approach to assessing shoreline change dynamics along the eThekwini coast of South Africa","authors":"Ekang C. Amatebelle,&nbsp;Zachariah H. Mshelia,&nbsp;Abiodun A. Ogundeji,&nbsp;Solomon T. Owolabi","doi":"10.1016/j.ejrs.2026.01.004","DOIUrl":"10.1016/j.ejrs.2026.01.004","url":null,"abstract":"<div><div>The eThekwini coast represents a critical hotspot of vulnerability to coastal hazards, particularly erosion and flooding. However, studies assessing historical shoreline change in eThekwini over the past four decades remain limited. The objective of this study was to assess and quantify shoreline change dynamics in eThekwini using geospatial techniques. Satellite acquired imagery from 1984 to 2024 were process in ArcGIS 10.8 using the Modified Normalised Difference Water Index (MNDWI) and the Digital Shoreline Analysis System (DSAS v6.0). The study quantify shoreline change using established statistical metrics i.e. Net Shoreline Movement (NSM), End Point Rate (EPR), Linear Regression Rate (LRR), and Weighted Linear Regression (WLR). The results show that 57.45 % of the shoreline was accretional while 42.55 % were erosional. However, the long-term accretional trend is likely to shift towards erosion, as the overall change rate of −1.03 m/yr consistently exceeds the ± 0.66 m/yr uncertainty threshold, reflecting the combined influence of urban development, climate change and sea level rise impacts on the coast. Accretion dominated Reg-1 and Reg-4, whereas Reg-2 and Reg-3 experienced statistically significant erosion beyond uncertainty. The erosional hotspots were largely concentrated near major rivers including Umgeni, uMdloti, and uMkhomazi river mouths. These patterns strongly correlate with the changes in land use, coastal infrastructures, damming, sand mining, and extreme weather events affecting the coastal city. These results provide actionable evidence for coastal risk management by identifying erosion hotspots, informing municipal setback lines, guiding risk-based land-use planning, and prioritising adaptation measures under South Africa’s Integrated Coastal Management framework (ICM Act No. 24 of 2008, amended Act No. 36 of 2014).Therefore, this study advocates a transformative approach that combine hybrid engineering and nature-based solutions to reduce human vulnerability and provide essential baseline data to support adaptive coastal management, and disaster preparedness to enhance coastal resilience in eThekwini.</div></div>","PeriodicalId":48539,"journal":{"name":"Egyptian Journal of Remote Sensing and Space Sciences","volume":"29 1","pages":"Pages 72-85"},"PeriodicalIF":4.1,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146037840","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Leveraging multi-agent deep reinforcement learning for autonomous galaxy image classification","authors":"Dhanvanth Reddy Yerramreddy ,&nbsp;Jayasurya Marasani ,&nbsp;Don S.","doi":"10.1016/j.ejrs.2026.03.001","DOIUrl":"10.1016/j.ejrs.2026.03.001","url":null,"abstract":"<div><div>Deep Reinforcement Learning (DRL) provides a general optimization framework that has achieved strong results across many decision-making problems. In this study, we investigate the practical use of representative DRL algorithms as <em>reward-optimized classifiers</em> for galaxy morphology recognition on the Galaxy Zoo dataset. Each agent receives a galaxy image as observation, predicts a class label as an action, and obtains an immediate reward (+1 for correct, <span><math><mrow><mo>−</mo><mn>1</mn></mrow></math></span> for incorrect); therefore, the learning problem corresponds to a single-step RL setting (contextual bandit/one-step MDP). To enhance robustness, we propose an ensemble approach using weighted majority voting, integrating predictions from multiple DRL agents into a multi-AI system. Our experiments on the Galaxy Zoo dataset shows that ensemble agent achieved highest accuracy of 98.55%, followed by DQN, RPPO with 98%, and PPO, A2C with 94%. Also, in order to evaluate the efficiency of DRL agents, a comparative study has been done with various deep learning models that are well established for galaxy image classification. We also report baseline results from standard deep-learning classifiers trained under the same data preprocessing and evaluation protocol. Our goal is to provide a transparent comparison and an accessible introduction to DRL-style optimization for image classification, while outlining sequential decision-making and continual/online learning extensions as future work</div></div>","PeriodicalId":48539,"journal":{"name":"Egyptian Journal of Remote Sensing and Space Sciences","volume":"29 1","pages":"Pages 223-233"},"PeriodicalIF":4.1,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147396759","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Landslide susceptibility mapping using tree-based machine learning classifiers and remote sensing derived conditioning factors: A case study of Chikmagalur District, Western Ghats, India","authors":"Babitha Ganesh ,&nbsp;Shweta Vincent ,&nbsp;Sameena Pathan ,&nbsp;Ganesh V. Bhat","doi":"10.1016/j.ejrs.2026.02.004","DOIUrl":"10.1016/j.ejrs.2026.02.004","url":null,"abstract":"&lt;div&gt;&lt;div&gt;Chikmagalur district of Karnataka state, situated within the boundaries of the Western Ghats is highly susceptible to landslides, especially during the monsoon season. Despite the recurring nature of these slope failures, limited research has been conducted to assess and mitigate the risk of landslides in the region. Existing studies often lack a comprehensive analysis of the triggering elements and rely on basic machine learning (ML) techniques, even though there are several advanced techniques that are being adopted across the world. A comprehensive dataset of the study area was prepared by integrating twenty different Landslide Conditioning Factors (LCFs) sourced from different remote sensing techniques and the information of 197 historical landslide events acquired from the Geological Survey of India (GSI). The 5-fold stratified cross validation method was applied to generate training and testing dataset in different iterations. Four different tree-based ML classifiers including Decision Trees (DT), Random Forest (RF), Extreme Gradient Boosting (XGBoost) and Categorical Boosting (CatBoost) were employed to prepare the models to predict the landslide prone areas of the district. These classifiers were specifically chosen because they have the capability to handle feature importance and do not require separate feature selection methods, which are often subjective and difficult to standardize. These ensemble models were then evaluated using different performance metrics that are generally used to evaluate classification models. CatBoost classifier exhibited superior performance, achieving an accuracy of 87.93%, with a precision of 0.85, recall of 0.913, F1-score of 0.88, and an AUC-ROC value of 0.95. Although the RF model also demonstrated strong and competitive performance across all the evaluation metrics, CatBoost was selected for the final preparation of landslide susceptibility map (LSM) due to its comparatively higher recall and AUC-ROC values, which are critical for reliably identifying landslide-prone areas. Consequently, the final LSM was generated using the CatBoost model. According to the LSM, approximately 20.53% of the total district falls within the range of high susceptibility with prediction probability values ranging from 0.6 to 1.0. A major portion, around 47.04% lies within moderate susceptibility zones, and the remaining percentage corresponds to places that are relatively safe from slope failures. Furthermore, the feature importance scores extracted from CatBoost model revealed that slope, rainfall, soil type and distance to road are the main factors that contribute to triggering slope failures in the study area. The application of reverse geocoding techniques on the final LSM indicated that, southwestern and southern taluks including Mudigere, Sringeri, border regions of Koppa and the southern parts of Chikmagalur exhibit a high concentration of landslide prone areas compared to other places. This map serves as a ","PeriodicalId":48539,"journal":{"name":"Egyptian Journal of Remote Sensing and Space Sciences","volume":"29 1","pages":"Pages 209-222"},"PeriodicalIF":4.1,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147396758","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}