Improving the Accuracy of Landslide Detection in "Off-site" Area by Machine Learning Model Portability Comparison: A Case Study of Jiuzhaigou Earthquake, China

The rising machine learning (ML) models have become the preferred way for landslide detection based on remote sensing images, but the performance of these models in a sample-free area are rarely concerned in many studies. In this study, we used a cross-validation method (training model in one area and validation in another) to compare the model portability of trained ML models applied in an “off-site” area, as a consideration of the landslide detection ability of these models in sample-free areas. We integrate nighttime light imagery, multi-seasonal optical Landsat time-series and digital elevation data, and we employed support vector machines (SVM), artificial neural networks (ANN) and random forest (RF) models to classify the satellite imagery and identify landslides. Samples of two scenarios generated from two subareas of the Jiuzhaigou disaster-stricken region are used for the cross-application and accuracy evaluation of three ML models. The results revealed that when the trained models are applied in areas outside those in which they were developed, the landslide identification accuracy of these three models has declined. Especially for the SVM and ANN models, the accuracy is greatly reduced and there appears a seriously imbalanced user’s and producer’s accuracy. However, although the performance of the RF model is lower than that of SVM and ANN models in their local area, the RF model exhibits stable portability, and retains the original performance and achieves a satisfactory balance between overestimation and underestimation in “off-site” areas. An additional validation from a new area proved that the landslide detection performance of the RF model with stable portability is higher than that of the SVM and ANN models in “off-site” areas. The results suggest that evaluating the model portability through cross-application can be a useful way to determine the most suitable model for landslide detection in “off-site” areas with a similar geographic environment to model development areas, so as to maximize the accuracy of landslide detection based on limited samples.

[1]  Y. You,et al.  Assessment of debris-flow potential dangers in the Jiuzhaigou Valley following the August 8, 2017, Jiuzhaigou earthquake, western China , 2019, Engineering Geology.

[2]  Mariana Belgiu,et al.  Random forest in remote sensing: A review of applications and future directions , 2016 .

[3]  Christos Polykretis,et al.  Comparison and evaluation of landslide susceptibility maps obtained from weight of evidence, logistic regression, and artificial neural network models , 2018, Natural Hazards.

[4]  S. Franklin,et al.  Detecting translational landslide scars using segmentation of Landsat ETM+ and DEM data in the northern Cascade Mountains, British Columbia , 2003 .

[5]  Birgit Kleinschmit,et al.  Automated derivation and spatio-temporal analysis of landslide properties in southern Kyrgyzstan , 2016, Natural Hazards.

[6]  Biswajeet Pradhan,et al.  Automatic landslide detection using Dempster–Shafer theory from LiDAR-derived data and orthophotos , 2017 .

[7]  Fang Chen,et al.  A new technique for landslide mapping from a large-scale remote sensed image: A case study of Central Nepal , 2017, Comput. Geosci..

[8]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[9]  Kang-Tsung Chang,et al.  Bayesian framework for mapping and classifying shallow landslides exploiting remote sensing and topographic data , 2013 .

[10]  Leonardo Santurri,et al.  Seasonal landslide mapping and estimation of landslide mobilization rates using aerial and satellite images , 2011 .

[11]  Saro Lee,et al.  Use of an artificial neural network for analysis of the susceptibility to landslides at Boun, Korea , 2003 .

[12]  Tapas Ranjan Martha,et al.  Segment Optimization and Data-Driven Thresholding for Knowledge-Based Landslide Detection by Object-Based Image Analysis , 2011, IEEE Transactions on Geoscience and Remote Sensing.

[13]  Yang Hong,et al.  Advances in landslide nowcasting: evaluation of a global and regional modeling approach , 2012, Environmental Earth Sciences.

[14]  André Stumpf,et al.  Object-oriented mapping of landslides using Random Forests , 2011 .

[15]  Mustafa Neamah Jebur,et al.  Spatial prediction of landslide hazard at the Luxi area (China) using support vector machines , 2015, Environmental Earth Sciences.

[16]  M. Rossi,et al.  Landslide volumes and landslide mobilization rates in Umbria, central Italy , 2009 .

[17]  Gang Chen,et al.  Identification of Forested Landslides Using LiDar Data, Object-based Image Analysis, and Machine Learning Algorithms , 2015, Remote. Sens..

[18]  Giorgos Mountrakis,et al.  A meta-analysis of remote sensing research on supervised pixel-based land-cover image classification processes: General guidelines for practitioners and future research , 2016 .

[19]  Sang Michael Xie,et al.  Combining satellite imagery and machine learning to predict poverty , 2016, Science.

[20]  P. Reichenbach,et al.  A review of statistically-based landslide susceptibility models , 2018 .

[21]  J. Nichol,et al.  Satellite remote sensing for detailed landslide inventories using change detection and image fusion , 2005 .

[22]  Fausto Guzzetti,et al.  Semi-automatic recognition and mapping of rainfall induced shallow landslides using optical satellite images , 2011 .

[23]  Abdulhamit Subasi,et al.  Congestive heart failure detection using random forest classifier , 2016, Comput. Methods Programs Biomed..

[24]  Fausto Guzzetti,et al.  Landslides triggered by the 23 November 2000 rainfall event in the Imperia Province, Western Liguria, Italy , 2004 .

[25]  Clemens Eisank,et al.  An object-based approach for semi-automated landslide change detection and attribution of changes to landslide classes in northern Taiwan , 2015, Earth Science Informatics.

[26]  F. Ren,et al.  Landslide susceptibility mapping using rough sets and back-propagation neural networks in the Three Gorges, China , 2013, Environmental Earth Sciences.

[27]  Masamu Aniya,et al.  LANDSLIDE HAZARD MAPPING AND THE APPLICATION OF GIS IN THE KULEKHANI WATERSHED, NEPAL , 1999 .

[28]  Junwei Han,et al.  Automatic landslide detection from remote-sensing imagery using a scene classification method based on BoVW and pLSA , 2013 .

[29]  J. Muller,et al.  Mapping regional economic activity from night-time light satellite imagery , 2006 .

[30]  Qian Wang,et al.  Integration of Information Theory, K-Means Cluster Analysis and the Logistic Regression Model for Landslide Susceptibility Mapping in the Three Gorges Area, China , 2017, Remote. Sens..

[31]  Brett A. Bryan,et al.  A novel algorithm for calculating transition potential in cellular automata models of land-use/cover change , 2019, Environ. Model. Softw..

[32]  F. Guzzetti,et al.  Landslide inventory maps: New tools for an old problem , 2012 .

[33]  Rasmus Fensholt,et al.  Detecting and monitoring long-term landslides in urbanized areas with nighttime light data and multi-seasonal Landsat imagery across Taiwan from 1998 to 2017 , 2019, Remote Sensing of Environment.

[34]  He Zhang,et al.  Inventory and Spatial Distribution of Landslides Triggered by the 8th August 2017 MW 6.5 Jiuzhaigou Earthquake, China , 2018, Journal of Earth Science.

[35]  I. Yilmaz.  Comparison of landslide susceptibility mapping methodologies for Koyulhisar, Turkey: conditional probability, logistic regression, artificial neural networks, and support vector machine , 2010 .

[36]  Birgit Kleinschmit,et al.  Automated Spatiotemporal Landslide Mapping over Large Areas Using RapidEye Time Series Data , 2014, Remote. Sens..

[37]  Tetsuya Kubota,et al.  Comparison of GIS-based landslide susceptibility models using frequency ratio, logistic regression, and artificial neural network in a tertiary region of Ambon, Indonesia , 2018, Geomorphology.

[38]  Stefan Lang,et al.  Object-based image analysis for remote sensing applications: modeling reality – dealing with complexity , 2008 .

[39]  Michele Dalponte,et al.  Tree Species Classification in Boreal Forests With Hyperspectral Data , 2013, IEEE Transactions on Geoscience and Remote Sensing.

[40]  Biswajeet Pradhan,et al.  Data Fusion Technique Using Wavelet Transform and Taguchi Methods for Automatic Landslide Detection From Airborne Laser Scanning Data and QuickBird Satellite Imagery , 2016, IEEE Transactions on Geoscience and Remote Sensing.

[41]  Biswajeet Pradhan,et al.  Spatial prediction models for shallow landslide hazards: a comparative assessment of the efficacy of support vector machines, artificial neural networks, kernel logistic regression, and logistic model tree , 2016, Landslides.

[42]  Biswajeet Pradhan,et al.  Modeling landslide susceptibility in data-scarce environments using optimized data mining and statistical methods , 2018 .

[43]  Eric Pirard,et al.  Automatic landslide detection from remote sensing images using supervised classification methods , 2007, 2007 IEEE International Geoscience and Remote Sensing Symposium.

[44]  Steven E. Franklin,et al.  A comparison of pixel-based and object-based image analysis with selected machine learning algorithms for the classification of agricultural landscapes using SPOT-5 HRG imagery , 2012 .

[45]  L. Hurni,et al.  Remote sensing of landslides: An analysis of the potential contribution to geo-spatial systems for hazard assessment in mountainous environments , 2005 .

[46]  Akansha Singh,et al.  Detection of 2011 Sikkim earthquake-induced landslides using neuro-fuzzy classifier and digital elevation model , 2016, Natural Hazards.

[47]  S. L. Kuriakose,et al.  Spatial data for landslide susceptibility, hazard, and vulnerability assessment: An overview , 2008 .

[48]  K. Lin,et al.  Distinguishing the windthrow and hydrogeological effects of typhoon impact on agricultural lands: an integrative OBIA and PPGIS approach , 2018 .

[49]  Jung Hyun Lee,et al.  A novel ensemble bivariate statistical evidential belief function with knowledge-based analytical hierarchy process and multivariate statistical logistic regression for landslide susceptibility mapping , 2014 .

[50]  B. Pradhan,et al.  Delineation of landslide hazard areas on Penang Island, Malaysia, by using frequency ratio, logistic regression, and artificial neural network models , 2010 .

[51]  Paolo Tarolli,et al.  Shallow Landslide Susceptibility Modeling Incorporating Rainfall Statistics: A Case Study from the Deokjeok-ri Watershed, South Korea , 2016 .

[52]  Cardona Alzate,et al.  Predicción y selección de variables con bosques aleatorios en presencia de variables correlacionadas , 2020 .

[53]  M. Marjanović,et al.  Landslide susceptibility assessment using SVM machine learning algorithm , 2011 .

[54]  Boris Schröder,et al.  How can statistical models help to determine driving factors of landslides , 2012 .

[55]  Thomas Oommen,et al.  A comparative analysis of pixel- and object-based detection of landslides from very high-resolution images , 2018, Int. J. Appl. Earth Obs. Geoinformation.

[56]  Yu Huang,et al.  Review on landslide susceptibility mapping using support vector machines , 2018, CATENA.

[57]  Kang-Tsung Chang,et al.  Combining spectral and geoenvironmental information for probabilistic event landslide mapping , 2014 .

[58]  A-Xing Zhu,et al.  A comparative study of an expert knowledge-based model and two data-driven models for landslide susceptibility mapping , 2018, CATENA.

[59]  Thomas Blaschke,et al.  Evaluation of Different Machine Learning Methods and Deep-Learning Convolutional Neural Networks for Landslide Detection , 2019, Remote. Sens..

[60]  Chia-Chi Chang,et al.  Modeling Typhoon Event-Induced Landslides Using GIS-Based Logistic Regression: A Case Study of Alishan Forestry Railway, Taiwan , 2013 .

[61]  Matthew C. Larsen,et al.  Assessing Landslide Hazards , 2007, Science.

[62]  Biswajeet Pradhan,et al.  Landslide susceptibility mapping at Hoa Binh province (Vietnam) using an adaptive neuro-fuzzy inference system and GIS , 2012, Comput. Geosci..

[63]  M. Hsu,et al.  Modeling typhoon- and earthquake-induced landslides in a mountainous watershed using logistic regression , 2007 .

[64]  Bahareh Kalantar,et al.  Assessment of the effects of training data selection on the landslide susceptibility mapping: a comparison between support vector machine (SVM), logistic regression (LR) and artificial neural networks (ANN) , 2018 .

[65]  Paul M. Mather,et al.  Support vector machines for classification in remote sensing , 2005 .

[66]  Weitao Chen,et al.  Forested landslide detection using LiDAR data and the random forest algorithm: A case study of the Three Gorges, China , 2014 .

[67]  D. Keefer,et al.  Investigating Landslides Caused by Earthquakes – A Historical Review , 2002 .