Comparative Study of Convolutional Neural Network and Conventional Machine Learning Methods for Landslide Susceptibility Mapping

Landslide susceptibility mapping (LSM) is a useful tool to estimate the probability of landslide occurrence, providing a scientific basis for natural hazards prevention, land use planning, and economic development in landslide-prone areas. To date, a large number of machine learning methods have been applied to LSM, and recently the advanced convolutional neural network (CNN) has been gradually adopted to enhance the prediction accuracy of LSM. The objective of this study is to introduce a CNN-based model in LSM and systematically compare its overall performance with the conventional machine learning models of random forest, logistic regression, and support vector machine. Herein, we selected Zhangzha Town in Sichuan Province, China, and Lantau Island in Hong Kong, China, as the study areas. Each landslide inventory and corresponding predisposing factors were stacked to form spatial datasets for LSM. The receiver operating characteristic analysis, area under the curve (AUC), and several statistical metrics, such as accuracy, root mean square error, Kappa coefficient, sensitivity, and specificity, were used to evaluate the performance of the models. Finally, the trained models were calculated, and the landslide susceptibility zones were mapped. Results suggest that both CNN and conventional machine learning-based models have a satisfactory performance. The CNN-based model exhibits an excellent prediction capability and achieves the highest performance but also significantly reduces the salt-of-pepper effect, which indicates its great potential for application to LSM.

A GIS-Based Landslide Susceptibility Mapping and Variable Importance Analysis Using Artificial Intelligent Training-Based Methods

Landslides often cause significant casualties and economic losses, and therefore landslide susceptibility mapping (LSM) has become increasingly urgent and important. The potential of deep learning (DL) like convolutional neural networks (CNN) based on landslide causative factors has not been fully explored yet. The main target of this study is the investigation of a GIS-based LSM in Zanjan, Iran and to explore the most important causative factor of landslides in the case study area. Different machine learning (ML) methods have been employed and compared to select the best results in the case study area. The CNN is compared with four ML algorithms, including random forest (RF), artificial neural network (ANN), support vector machine (SVM), and logistic regression (LR). To do so, sixteen landslide causative factors have been extracted and their related spatial layers have been prepared. Then, the algorithms were trained with related landslide and non-landslide points. The results illustrate that the five ML algorithms performed suitably (precision = 82.43–85.6%, AUC = 0.934–0.967). The RF algorithm achieves the best result, while the CNN, SVM, the ANN, and the LR have the best results after RF, respectively, in this case study. Moreover, variable importance analysis results indicate that slope and topographic curvature contribute more to the prediction. The results would be beneficial to planning strategies for landslide risk management.

Integration of Information Theory, Fractal Theory and Statistical Analyses for Landslide Susceptibility Mapping at Regional Scales

A new model, integrating information theory, fractal theory and statistical model for accurate landslide susceptibility mapping (LSM) at regional scales, has been proposed. In this model, landslide conditional factors are firstly classified with an optimal number of classes, which is determined by maximizing their information coefficients estimated from Shannon’s entropy model. The spatial association between influencing factors and induced landslides has been measured by introducing the variable fractal dimension method (VFDM). The VFDM approach fully considers the characteristics of landslide fractal distribution. Then the fractal dimensions (\(D\)) are calculated to provide multiple factors with various numerical weights. The proposed model eventually combines the landslide frequency ratio (\(fr\)) of each factor with corresponding weight to achieve spatial prediction of landslides, illustrated by an example area in China. In the study area, 500 landslides have been identified by aerial photograph interpretation, extensive field investigations, historical and bibliographical landslide data. In the model, these landslides are randomly split into a training dataset (70 %)and a validating dataset (30 %) Seven factors are recognized and analyzed by frequency ratio (FR) method, including lithology, distance to fault, altitude, slope, aspect, distance to stream and distance to the road. The receiver operating characteristic curve (AUROC) has been adopted to compare and validate the model results. Results show that the proposed landslide model achieved a more accurate prediction with AUROC equal to 0.8467, over-performing than the conventional frequency ratio method (AUROC=0.8088). According to the final prognostic landslide susceptibility map, 16.37 % f the study area shows very high and high susceptibility, accounting for 63.55 % f the entire landslides. Evaluation of relative factor importance based on a one-by-one factor removal test indicates that the lithology factor contributes unique information for landslides. In conclusion, the example demonstrates that the proposed framework is promising for further improvement of LSM.