Application of Machine Learning Algorithms and Their Ensemble for Landslide Susceptibility Mapping

Kalantar, Bahareh; Saeidi, Vahideh; Ahmadi, Parisa

doi:10.1007/978-3-030-60227-7_25

Cited by 6 publications

(2 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Due to improvement of spatial and temporal resolution of satellite imagery and availability of synthetic aperture radar (SAR) dataset, disaster mapping based on RS data has been converted into a hot topic [33,34]. Meanwhile, algorithm development and computer sciences revolutionized the accuracy and time of different sensors' data processing in various fields such as RS, prediction of natural disaster, feature detection, and biomedical [35,36]. The machine learning (ML) algorithms are among the most popular methods in the image processing and computer vision.…”

Section: Introductionmentioning

confidence: 99%

Fusion of the Multisource Datasets for Flood Extent Mapping Based on Ensemble Convolutional Neural Network (CNN) Model

Seydi

Saeidi²,

Kalantar

et al. 2022

Journal of Sensors

View full text Add to dashboard Cite

Floods, as one of the natural hazards, can affect the environment, damage the infrastructures, and threaten human lives. Due to climate change and anthropogenic activities, floods occur in high frequency all over the world. Therefore, mapping of the flood areas is of prime importance in disaster management. This research presents a novel framework for flood area mapping based on heterogeneous remote sensing (RS) datasets. The proposed framework fuses the synthetic aperture radar (SAR), optical, and altimetry datasets for mapping flood areas, and it is applied in three main steps: (1) preprocessing, (2) deep feature extraction based on multiscale residual kernel convolution and convolution neural network’s (CNN) parameter optimization by fusing the datasets, and (3) flood detection based on the trained model. This research exploits two large-scale area datasets for mapping the flooded areas in Golestan and Khuzestan provinces, Iran. The results show that the proposed methodology has a high performance in flood area detection. The visual and numerical analyses verify the effectiveness and ability of the proposed method to detect the flood areas with an overall accuracy (OA) higher than 98% in both study areas. Finally, the efficiency of the designed architecture was verified by hybrid-CNN and 3D-CNN methods.

show abstract

Section: Introductionmentioning

confidence: 99%

Fusion of the Multisource Datasets for Flood Extent Mapping Based on Ensemble Convolutional Neural Network (CNN) Model

Seydi

Saeidi²,

Kalantar

et al. 2022

Journal of Sensors

View full text Add to dashboard Cite

show abstract

“…Based on the scale of study these approaches are applied in all scale ranges (Aleotti and Chowdhury 1999;Shano et al 2020). The quantitative approach is grouped into different categories such as deterministic or mechanical methods that were applied by various researchers (e.g., Sarkar et al 2016;Ciurleo et al 2017;Krušić et al 2017;Shano et al 2020) and statistical (e.g., Carrara et al 1991;Jager and Wieczorek 1994;Donati and Turrini 2002;Zhou et al 2002;Meten et al 2015;Berhane et al 2020) and machine learning/artificial intelligent (Catani et al 2005;Pradhan 2013;Ada and San 2017;Kavzoglu et al 2018;Hu et al 2020;Kalantar et al 2020a;Shano et al 2020;Wang et al 2020). Some of the researchers also used hybrid types of approaches (artificial intelligence with statistical methods) for landslide analysis (Chen et al 2018;Li and Chen 2019;Nguyen et al 2019).…”

Section: Introductionmentioning

confidence: 99%

Landslide susceptibility mapping using frequency ratio model: the case of Gamo highland, South Ethiopia

2021

View full text Add to dashboard Cite

Landslide is the most frequently occurring geo-hazard in mountainous terrains of the world. It affects human life, infrastructures, landscapes, and human properties as well as their day-to-day activities. In the current study area which is found in the Gamo Zone of south Ethiopia, recurrent landslide hazards have occurred. To minimize this landslide hazard on human life and their properties, landslide susceptibility mapping is an important step for environmental planning. For this purpose, 1554 landslides and 9 landslide causative factors (both conditioning and triggering factors) were used. Each thematic layer has different classes and some of these classes influence landslide occurrence more than others. The most influencing factor classes which were identified by the frequency ratio model include slope classes between12 and 45°; convex and concave classes of the curvature; aspect classes of north, northeast, south, and southwest directions; and elevation classes in between 2118 and 2492 m. The distances factors, proximity to streams, and lineaments 0-100 m and 0-200 m respectively have a very high on landslide occurrences. Land use/land cover factor has different classes and they have different levels of direct and indirect influences on landslide occurrences. The landslide susceptibility map was classified as very low, low, moderate, high, and very high classes each accounting for 17.8%, 29.19%, 28.55%, 17.52%, and 6.91% of the area respectively. To evaluate the reliability of this model, the landslide susceptibility map was verified using a receiver operating characteristic (ROC) curve with a value of 82% and through field observation. Therefore, this can be used by local, zonal, regional, and federal governments for land use planning, disaster prevention, and mitigation as it offers first-hand information.

show abstract

Properties of fault zones and their influences on rainfall-induced landslides, examples from Alborz and Zagros ranges

Ehteshami‐Moinabadi

2022

Environ Earth Sci

View full text Add to dashboard Cite

Application of Machine Learning Algorithms and Their Ensemble for Landslide Susceptibility Mapping

Cited by 6 publications

References 21 publications

Fusion of the Multisource Datasets for Flood Extent Mapping Based on Ensemble Convolutional Neural Network (CNN) Model

Fusion of the Multisource Datasets for Flood Extent Mapping Based on Ensemble Convolutional Neural Network (CNN) Model

Landslide susceptibility mapping using frequency ratio model: the case of Gamo highland, South Ethiopia

Properties of fault zones and their influences on rainfall-induced landslides, examples from Alborz and Zagros ranges

Contact Info

Product

Resources

About