☆ 4.7 Article

Rapid Landslide Extraction from High-Resolution Remote Sensing Images Using SHAP-OPT-XGBoost

REMOTE SENSING (2023)

Related references

Note: Only part of the references are listed.

Article Environmental Sciences

Comparisons of Convolutional Neural Network and Other Machine Learning Methods in Landslide Susceptibility Assessment: A Case Study in Pingwu

Ziyu Jiang et al.

Summary: This study compares the characteristics and effectiveness of convolutional neural network (CNN) and conventional machine learning (ML) methods in landslide susceptibility assessment (LSA). The result shows that CNN achieves the highest accuracy (86.41%) and the highest AUC (0.9249) in LSA. The CNN model performs better in identifying landslide cluster regions due to its consideration of surrounding environment information through convolution operation. The high accuracy and concentrated possibility of CNN in LSA is of great significance for disaster prevention and mitigation, although there is room for improvement in identifying low-cluster landslide areas.

REMOTE SENSING (2023)

Add to Collection

Article Geography, Physical

Assessing the effectiveness of alternative landslide partitioning in machine learning methods for landslide prediction in the complex Himalayan terrain

Muhammad Tayyib Riaz et al.

Summary: This study analyzes the effectiveness of different landslide partitioning techniques in landslide susceptibility modelling in Muzaffarabad, Pakistan. The results show that using cluster training samples leads to more accurate and reliable models for landslide susceptibility.

PROGRESS IN PHYSICAL GEOGRAPHY-EARTH AND ENVIRONMENT (2023)

Add to Collection

Article Environmental Sciences

Insights into geospatial heterogeneity of landslide susceptibility based on the SHAP-XGBoost model

Junyi Zhang et al.

Summary: The study aimed to construct an explanatory framework for landslide susceptibility evaluation models based on the SHAP-XGBoost algorithm. By selecting typical mountainous regions, a geospatial database containing 12 influencing factors was constructed, and the prediction results of the model were explained using the SHAP algorithm. The study found that the spatial distribution of landslides is heterogeneous and complex, and the contribution of each influencing factor has regional characteristics and spatial heterogeneity. The generalizability of the landslide susceptibility evaluation model is spatially heterogeneous, with better generalizability to regions with similar characteristics.

JOURNAL OF ENVIRONMENTAL MANAGEMENT (2023)

Add to Collection

Article Geosciences, Multidisciplinary

A LightGBM-based landslide susceptibility model considering the uncertainty of non-landslide samples

Deliang Sun et al.

Summary: This article aims to construct a data-driven landslide susceptibility model that takes into account the selection of non-landslide samples. The models were established by selecting conditioning factors and grid units based on historical landslide events. The result showed that using the information quantity method improved model accuracy and identified high susceptibility areas. The study emphasizes the importance of sample selection in a binary classification model.

GEOMATICS NATURAL HAZARDS & RISK (2023)

Add to Collection

Article Environmental Sciences

Assessment of a New Solar Radiation Nowcasting Method Based on FY-4A Satellite Imagery, the McClear Model and SHapley Additive exPlanations (SHAP)

Dongyu Jia et al.

Summary: The global warming effect is accelerating and poses a threat to human survival and health. Reliable renewable energy is the top priority in solving this problem. This study proposes a fast and accurate solar radiation nowcasting method using FY-4A satellite data and the McClear clear sky model. The results show that the random forest (RF) model performs better than the support vector regression (SVR) model, providing more accurate and stable radiation predictions, especially when the time horizon exceeds 60 minutes. This new method is important for radiation prediction in cities with limited meteorological observation and can guide the location planning of photovoltaic power stations.

REMOTE SENSING (2023)

Add to Collection

Article Forestry

Pine wilt disease detection in high-resolution UAV images using object-oriented classification

Zhao Sun et al.

Summary: This study used an unmanned aerial vehicle platform and multi-scale segmentation algorithm combined with object-oriented classification to successfully identify pine trees affected by pine wilt disease, providing a technical method for monitoring disease occurrence and identifying affected trees.

JOURNAL OF FORESTRY RESEARCH (2022)

Add to Collection

Article Engineering, Environmental

Advanced hyperparameter optimization for improved spatial prediction of shallow landslides using extreme gradient boosting (XGBoost)

Taskin Kavzoglu et al.

Summary: This study investigates the effectiveness and robustness of optimization algorithms in improving the performance of machine learning algorithms for landslide susceptibility mapping. The optimized models showed significant improvement in overall accuracy compared to the default setting, with GA and Hyperband methods performing the best in terms of predictive abilities.

BULLETIN OF ENGINEERING GEOLOGY AND THE ENVIRONMENT (2022)

Add to Collection

Article Environmental Sciences

Convolutional neural network (CNN) with metaheuristic optimization algorithms for landslide susceptibility mapping in Icheon, South Korea

Wahyu Luqmanul Hakim et al.

Summary: This study explores the use of deep learning algorithms in combination with the grey wolf optimizer and imperialist competitive algorithm for landslide susceptibility mapping, leading to improved accuracy compared to standalone CNN models. The optimized models outperformed previous research and generated more suitable models for landslide assessment in the study area.

JOURNAL OF ENVIRONMENTAL MANAGEMENT (2022)

Add to Collection

Article Chemistry, Analytical

Landslide Susceptibility Mapping Using Machine Learning Algorithm Validated by Persistent Scatterer In-SAR Technique

Muhammad Afaq Hussain et al.

Summary: The present research aims to identify landslide susceptibility along Karakoram Highway using LSM method and various models. PS-InSAR technology is used to evaluate slope deformation velocity and improve the accuracy of LSM. The results show that RF model combined with PS-InSAR outcomes yields superior findings and provides a new landslide susceptibility map for the study region, which can help mitigate landslide catastrophes and ensure roadway safety.

SENSORS (2022)

Add to Collection

Article Environmental Sciences

Landslide Extraction Using Mask R-CNN with Background-Enhancement Method

Ruilin Yang et al.

Summary: The application of deep learning methods has improved the accuracy and automation of landslide extractions based on remote sensing images. However, the quality of training samples often fails the requirement for training deep networks, and the similarity between background objects and landslides can lead to false and missed extractions. To solve these problems, a background-enhancement method was proposed to enrich the complexity of samples and landslide-inducing attributes were used as supplements. The experiments showed a significant improvement in extraction accuracy using the proposed methods.

REMOTE SENSING (2022)

Add to Collection

Article Geography, Physical

Evaluation of neural network models for landslide susceptibility assessment

Yaning Yi et al.

Summary: This study evaluated and compared three advanced neural network models for landslide susceptibility assessment and found that all three models performed satisfactorily in accurately predicting susceptible regions. The RNN model outperformed the other two models in terms of accuracy but had higher complexity.

INTERNATIONAL JOURNAL OF DIGITAL EARTH (2022)

Add to Collection

Article Environmental Sciences

Earthquake-Induced Landslide Susceptibility Assessment Using a Novel Model Based on Gradient Boosting Machine Learning and Class Balancing Methods

Shuhao Zhang et al.

Summary: This paper proposes an advanced AI model based on the DCE loss function and XGBoost or LGB to predict the susceptibility of a specific part of a landslide. The proposed method improves the class imbalance issue and produces more accurate results compared to existing models. The XGB(DCE) model with DCE loss achieves the best prediction performance.

REMOTE SENSING (2022)

Add to Collection

Article Engineering, Geological

Metaheuristic-based support vector regression for landslide displacement prediction: a comparative study

Junwei Ma et al.

Summary: This study proposes a hybrid approach integrating k-fold cross validation, metaheuristic support vector regression, and the nonparametric Friedman test to enhance reproducibility in landslide displacement prediction. The results demonstrate that the hybrid approach improves accuracy and reliability in ML-based prediction and recommends the use of PSO algorithm for hyperparameter tuning.

LANDSLIDES (2022)

Add to Collection

Article Engineering, Geological

Rapid prediction of landslide dam stability considering the missing data using XGBoost algorithm

Ning Shi et al.

Summary: In this study, a stability prediction method for landslide dams based on the XGBoost model was established, and a landslide dam database containing 2783 historical cases from 49 countries worldwide was created. The results show that the XGBoost prediction model can predict landslide dam stability even with missing data, and its accuracy is improved compared to traditional rapid evaluation methods. Furthermore, a characteristic analysis framework based on the SHapley Additive exPlanations attribution theory was proposed, demonstrating the importance and influence of each characteristic on landslide dam stability.

LANDSLIDES (2022)

Add to Collection

Article Forestry

Automatic Remote Sensing Identification of Co-Seismic Landslides Using Deep Learning Methods

Dongdong Pang et al.

Summary: This paper establishes a convolutional neural network model using the YOLOv3 algorithm to automatically recognize co-seismic landslides in satellite remote sensing images, demonstrating high recognition accuracy and fast speed.

FORESTS (2022)

Add to Collection

Article Environmental Sciences

Exploring aspects affecting the predicted capacity of landslide susceptibility based on machine learning technology

Qiang Liu et al.

Summary: This study investigates the causes of prediction uncertainty in landslide susceptibility using machine learning. Various factors and models were analyzed, and the study area was found to have concentrated landslides in the central and southeastern parts. The evaluation of different machine learning algorithms led to the identification of the optimal model.

GEOCARTO INTERNATIONAL (2022)

Add to Collection

Article Geosciences, Multidisciplinary

Explainable step-wise binary classification for the susceptibility assessment of geo-hydrological hazards

Omer Ekmekcioglu et al.

Summary: This research introduces a novel step-wise binary prediction framework for assessing the susceptibility of geo-hydrological hazards, specifically floods and landslides. The framework consists of two major steps: predicting hazard-prone locations and classifying floods and landslides separately. The study utilizes historically experienced hazard locations and hazard conditioning factors to provide accurate predictions. The results demonstrate a strong agreement between the predicted and observed hazard locations, highlighting the effectiveness of the proposed hybrid prediction framework.

CATENA (2022)

Add to Collection

Review Chemistry, Multidisciplinary

Deep Residual Learning for Image Recognition: A Survey

Muhammad Shafiq et al.

Summary: This survey introduces the advantages of deep residual networks on ImageNet and their implications for future research. It also discusses the significant advancements and unresolved issues in the implementation of deep residual networks in practice.

APPLIED SCIENCES-BASEL (2022)

Add to Collection

Article Multidisciplinary Sciences

Predictive Performances of Ensemble Machine Learning Algorithms in Landslide Susceptibility Mapping Using Random Forest, Extreme Gradient Boosting (XGBoost) and Natural Gradient Boosting (NGBoost)

Taskin Kavzoglu et al.

Summary: In this study, a novel ensemble learning method called NGBoost was proposed for landslide susceptibility modeling in Turkey. The predictive performance of NGBoost was compared to other ensemble learning methods, and the results showed that NGBoost had the highest predictive ability and outperformed other methods by approximately 6% in terms of overall accuracy.

ARABIAN JOURNAL FOR SCIENCE AND ENGINEERING (2022)

Add to Collection

Article Geosciences, Multidisciplinary

Landslide identification using machine learning

Haojie Wang et al.

Summary: This paper proposes a novel machine-learning and deep-learning method to identify natural-terrain landslides using integrated geodatabases. The method was demonstrated to have high accuracy in a case study, with CNN achieving the highest identification accuracy. By utilizing machine learning and deep learning techniques, the proposed method shows outstanding robustness and great potential in tackling the landslide identification problem.

GEOSCIENCE FRONTIERS (2021)

Add to Collection

Article Computer Science, Artificial Intelligence

A comparative analysis of gradient boosting algorithms

Candice Bentejac et al.

Summary: The family of gradient boosting algorithms has been expanded with XGBoost, LightGBM, and CatBoost, which focus on reliability, efficiency, speed, and accuracy. In the comparison study, CatBoost is the best in generalization accuracy and AUC, LightGBM is the fastest but not the most accurate, and XGBoost ranks second in accuracy and training speed.

ARTIFICIAL INTELLIGENCE REVIEW (2021)

Add to Collection

Article Environmental Sciences

Hybrids of Support Vector Regression with Grey Wolf Optimizer and Firefly Algorithm for Spatial Prediction of Landslide Susceptibility

Ru Liu et al.

Summary: This study evaluates landslide susceptibility in Zhenping County using a hybrid model, combining SVR with GWO and FA, based on a database of landslides and various conditioning factors. The results show that the SVR-GWO model performs the best in spatial prediction of landslides, followed by SVR-FA and SVR models. The hybrid models improve the performance of the single SVR model and show good prospects for regional-scale landslide spatial modeling.

REMOTE SENSING (2021)

Add to Collection

Article Environmental Sciences

A Meta-Learning Approach of Optimisation for Spatial Prediction of Landslides

Biswajeet Pradhan et al.

Summary: The study presents an optimization method based on meta-learning for spatial prediction of landslides, which starts with promising configurations based on basic and statistical meta-features. Compared to the Bayesian method, this approach achieved slightly lower accuracies on the training and test datasets.

REMOTE SENSING (2021)

Add to Collection

Article Environmental Sciences

Comparison of multi-criteria and artificial intelligence models for land-subsidence susceptibility zonation

Alireza Arabameri et al.

Summary: This study aimed to predict land subsidence susceptibility in the Shahroud plain, Iran, by using MCDM and AI models. Among the different methods used, the Cforest model showed the best performance, identifying 30.55% of the study area as extremely vulnerable to land subsidence. The results of this research will be valuable for planners and policymakers in implementing appropriate development strategies in the region.

JOURNAL OF ENVIRONMENTAL MANAGEMENT (2021)

Add to Collection

Article Engineering, Electrical & Electronic

Landslide Detection Mapping Employing CNN, ResNet, and DenseNet in the Three Gorges Reservoir, China

Tong Liu et al.

Summary: In this study, an accurate Landslide Detection Mapping (LDM) model was constructed based on convolutional neural networks, residual neural networks, and DenseNets, considering ZY-3 high spatial resolution data and conditioning factors. The experimental results demonstrated that these models performed well, with accuracy above 0.95. DenseNet incorporating RS images and CFs outperformed other models in terms of evaluation metrics, with improvements in Kappa coefficients and ACC. Elevation factor was identified as the most important factor in the landslide model construction experiment.

IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING (2021)

Add to Collection

Article Geosciences, Multidisciplinary

A comparison among fuzzy multi-criteria decision making, bivariate, multivariate and machine learning models in landslide susceptibility mapping

Quoc Bao Pham et al.

Summary: This study analyzed the landslide susceptibility in the Kysuca river basin in Slovakia using various models, with RFC model proving to be the most accurate. Additionally, the FDEMATEL-ANP model demonstrated better prediction capability compared to multivariate LR, machine learning NBC, and deep learning XGBoost models.

GEOMATICS NATURAL HAZARDS & RISK (2021)

Add to Collection

Article Geosciences, Multidisciplinary

Application of alternating decision tree with AdaBoost and bagging ensembles for landslide susceptibility mapping

Yanli Wu et al.

CATENA (2020)

Add to Collection

Article Environmental Sciences

Parallel Regional Segmentation Method of High-Resolution Remote Sensing Image Based on Minimum Spanning Tree

Wenjie Lin et al.

REMOTE SENSING (2020)

Add to Collection

Article Environmental Sciences

Landslides Information Extraction Using Object-Oriented Image Analysis Paradigm Based on Deep Learning and Transfer Learning

Heng Lu et al.

REMOTE SENSING (2020)

Add to Collection

Article Mathematics

Predicting Hard Rock Pillar Stability Using GBDT, XGBoost, and LightGBM Algorithms

Weizhang Liang et al.

MATHEMATICS (2020)

Add to Collection

Article Computer Science, Interdisciplinary Applications

Integration of convolutional neural network and conventional machine learning classifiers for landslide susceptibility mapping

Zhice Fang et al.

COMPUTERS & GEOSCIENCES (2020)

Add to Collection

Article Environmental Sciences

Landslide Susceptibility Mapping Using Machine Learning Algorithms and Remote Sensing Data in a Tropical Environment

Viet-Ha Nhu et al.

INTERNATIONAL JOURNAL OF ENVIRONMENTAL RESEARCH AND PUBLIC HEALTH (2020)

Add to Collection

Article Environmental Sciences

Novel Credal Decision Tree-Based Ensemble Approaches for Predicting the Landslide Susceptibility

Alireza Arabameri et al.

REMOTE SENSING (2020)

Add to Collection

Article Environmental Sciences

Rainfall Induced Landslide Susceptibility Mapping Based on Bayesian Optimized Random Forest and Gradient Boosting Decision Tree Models-A Case Study of Shuicheng County, China

Guangzhi Rong et al.

WATER (2020)

Add to Collection

Article Green & Sustainable Science & Technology

Landslides Susceptibility Assessment Based on GIS Statistical Bivariate Analysis in the Hills Surrounding a Metropolitan Area

Paul Sestras et al.

SUSTAINABILITY (2019)

Add to Collection

Article Environmental Sciences

Detecting and monitoring long-term landslides in urbanized areas with nighttime light data and multi-seasonal Landsat imagery across Taiwan from 1998 to 2017

Tzu-Hsin Karen Chen et al.

REMOTE SENSING OF ENVIRONMENT (2019)

Add to Collection

Article Environmental Sciences

Debris Flow Susceptibility Mapping Using Machine-Learning Techniques in Shigatse Area, China

Yonghong Zhang et al.

REMOTE SENSING (2019)

Add to Collection

Article Engineering, Geological

The new landslide inventory of Tuscany (Italy) updated with PS-InSAR: geomorphological features and landslide distribution

A. Rosi et al.

LANDSLIDES (2018)

Add to Collection

Review Geosciences, Multidisciplinary

Review on landslide susceptibility mapping using support vector machines

Yu Huang et al.

CATENA (2018)

Add to Collection

Article Environmental Sciences

An Efficient Parallel Multi-Scale Segmentation Method for Remote Sensing Imagery

Haiyan Gu et al.

REMOTE SENSING (2018)

Add to Collection

Review Environmental Sciences

Remote Sensing of LandslidesA Review

Chaoying Zhao et al.

REMOTE SENSING (2018)

Add to Collection

Article Geography, Physical

Efficient paddy field mapping using Landsat-8 imagery and object-based image analysis based on advanced fractel net evolution approach

Tengfei Su

GISCIENCE & REMOTE SENSING (2017)

Add to Collection

Article Engineering, Civil

A comparative assessment of GIS-based data mining models and a novel ensemble model in groundwater well potential mapping

Seyed Amir Naghibi et al.

JOURNAL OF HYDROLOGY (2017)

Add to Collection

Article Engineering, Geological

Landslide dam formation susceptibility analysis based on geomorphic features

Chien-Yuan Chen et al.

LANDSLIDES (2016)

Add to Collection

Article Geography, Physical

GIS-based logistic regression for landslide susceptibility mapping of the Zhongxian segment in the Three Gorges area, China

Shi-Biao Bai et al.

GEOMORPHOLOGY (2010)

Add to Collection

Article Computer Science, Information Systems

ESP: a tool to estimate scale parameter for multiresolution image segmentation of remotely sensed data

Lucian Dragut et al.

INTERNATIONAL JOURNAL OF GEOGRAPHICAL INFORMATION SCIENCE (2010)

Add to Collection

Article Statistics & Probability

Greedy function approximation: A gradient boosting machine

JH Friedman

ANNALS OF STATISTICS (2001)

Add to Collection

© Peeref 2019-2024. All rights reserved.