Evaluation of deep learning algorithms for national scale landslide susceptibility mapping of Iran

Ngo, Phuong Thao Thi; Panahi, Mahdi; Khosravi, Khabat; Ghorbanzadeh, Omid; Kariminejad, Narges; Cerdà, Artemi; Lee, Saro

doi:10.1016/j.gsf.2020.06.013

Cited by 265 publications

(65 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Such multi-temporal seasonal and historic inventories are typically developed using satellite-based monitoring of a given region (e.g., Behling et al, 2016). Landslide susceptibility models are commonly applied across a range of spatial scales (Cascini, 2008), from individual slope units (e.g., Alvioli et al, 2016;Amato et al, 2019) to catchments (e.g., Conforti et al, 2012;Romer & Ferentinou, 2016) to geographical regions (e.g., Sabatakakis et al, 2013;Thi Ngo et al, 2020) and even globally (e.g., L. Lin et al, 2017;Stanley & Kirschbaum, 2017). Statistical landslide susceptibility modeling is very common (Reichenbach et al, 2018), and is often a fundamental component of landslide hazard analyses, risk assessments, land-use planning, and early warning systems (e.g., Fell et al, 2008;Palau et al, 2020;van Westen et al, 2008).…”

mentioning

confidence: 99%

Temporal Variations in Landslide Distributions Following Extreme Events: Implications for Landslide Susceptibility Modeling

et al. 2021

View full text Add to dashboard Cite

Landslides are globally occurring natural hazards posing significant threats to life and sustainable development (Froude & Petley, 2018). Statistics from the Center for Research on the Epidemiology of Disasters reveal that landslides account for 17% of all fatalities due to natural hazards (Sassa & Canuti, 2009), whilst World Bank research suggests >66 million people live in high risk landslide regions (Dilley, 2005). Such statistics highlight a compelling need for efforts to manage and mitigate landslide risk with constant evaluation and improvement.A fundamental component of landslide management strategies are landslide susceptibility models, which predict the likely geographic locations of future landslides (Guzzetti et al., 2006). Landslide susceptibility models can be developed using physically based techniques (e.g.,

show abstract

mentioning

confidence: 99%

Temporal Variations in Landslide Distributions Following Extreme Events: Implications for Landslide Susceptibility Modeling

et al. 2021

View full text Add to dashboard Cite

show abstract

“…CNN-based deep learning can also be used to generate spatial probabilistic models. For example, Gagne et al [76] explored DL for the probabilistic prediction of severe hailstorms, while Thi Ngo et al [77] assessed landslide susceptibility modeling. If the primary output will be a probabilistic model, probabilistic-based assessment metrics should be reported.…”

Section: Semantic Segmentationmentioning

confidence: 99%

Accuracy Assessment in Convolutional Neural Network-Based Deep Learning Remote Sensing Studies—Part 2: Recommendations and Best Practices

2021

View full text Add to dashboard Cite

Convolutional neural network (CNN)-based deep learning (DL) has a wide variety of applications in the geospatial and remote sensing (RS) sciences, and consequently has been a focus of many recent studies. However, a review of accuracy assessment methods used in recently published RS DL studies, focusing on scene classification, object detection, semantic segmentation, and instance segmentation, indicates that RS DL papers appear to follow an accuracy assessment approach that diverges from that of traditional RS studies. Papers reporting on RS DL studies have largely abandoned traditional RS accuracy assessment terminology; they rarely reported a complete confusion matrix; and sampling designs and analysis protocols generally did not provide a population-based confusion matrix, in which the table entries are estimates of the probabilities of occurrence of the mapped landscape. These issues indicate the need for the RS community to develop guidance on best practices for accuracy assessment for CNN-based DL thematic mapping and object detection. As a first step in that process, we explore key issues, including the observation that accuracy assessments should not be biased by the CNN-based training and inference processes that rely on image chips. Furthermore, accuracy assessments should be consistent with prior recommendations and standards in the field, should support the estimation of a population confusion matrix, and should allow for assessment of model generalization. This paper draws from our review of the RS DL literature and the rich record of traditional remote sensing accuracy assessment research while considering the unique nature of CNN-based deep learning to propose accuracy assessment best practices that use appropriate sampling methods, training and validation data partitioning, assessment metrics, and reporting standards.

show abstract

“…Susceptibility prediction is an effective solution for dealing with environmental threats (Sun et al 2020, Wei et al 2021). In the case of landslides, prediction-oriented efforts are of great value to engineers and decision-makers toward providing appropriate mitigation measures and land use planning (Ngo et al 2021). Up to now, a wide variety of modeling tools and strategies have been proposed to model the susceptibility of landslide all over the world.…”

Section: Problem and Solutionmentioning

confidence: 99%

Landslide Susceptibility Zonation Using Statistical and Machine Learning Approaches in Northern Lecco, Italy

Mehrabi

2021

Preprint

View full text Add to dashboard Cite

This study deals with landslide susceptibility mapping in the northern part of Lecco Province, Lombardy Region, Italy. In so doing, a valid landslide inventory map and thirteen conditioning factors (including elevation, slope aspect, slope degree, plan curvature, profile curvature, distance to waterway, distance to road, distance to fault, soil type, land use, lithology, stream power index, and topographic wetness index) form the spatial database within geographic information system (GIS). The used evaluative models comprise a bivariate statistical approach called frequency ratio (FR) and two machine learning tools, namely multi-layer perceptron neural network (MLPNN) and adaptive neuro-fuzzy inference system (ANFIS). These models first use landslide and non-landslide records for comprehending the relationship between the landslide occurrence and conditioning factors. Then landslide susceptibility values are predicted for the whole area. The accuracy of the produced susceptibility maps is measured using area under the curve (AUC) index, according to which, the MLPNN (AUC = 0.916) presented the most accurate map, followed by the FR (AUC = 0. 898) and ANFIS (AUC = 0.889). Visual interpretation of the susceptibility maps, FR-based correlation analysis, as well as the importance assessment of conditioning factors, all indicated the significant contribution of the road networks to the crucial susceptibility of landslide. Lastly, an explicit predictive formula is extracted from the implemented MLPNN model for a convenient approximation of landslide susceptibility value.

show abstract

Evaluation of deep learning algorithms for national scale landslide susceptibility mapping of Iran

Cited by 265 publications

References 46 publications

Temporal Variations in Landslide Distributions Following Extreme Events: Implications for Landslide Susceptibility Modeling

Temporal Variations in Landslide Distributions Following Extreme Events: Implications for Landslide Susceptibility Modeling

Accuracy Assessment in Convolutional Neural Network-Based Deep Learning Remote Sensing Studies—Part 2: Recommendations and Best Practices

Landslide Susceptibility Zonation Using Statistical and Machine Learning Approaches in Northern Lecco, Italy

Contact Info

Product

Resources

About