GIS-based spatial prediction of landslide using road factors and random forest for Sichuan-Tibet Highway

Ye, Chengming; Wei, Ruilong; Ge, Yong; Li, Yao; Marcato, José; Li, Jonathan

doi:10.1007/s11629-021-6848-6

Cited by 22 publications

(14 citation statements)

References 68 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is effortless to derive that the new Belief Jensen-Renyi divergence can manage the limited symmetry of the two belief Renyi divergences, so the next theorem is provided without proof. This theorem points out that the new divergence is rearrangement invariant regarding ω within α's full definition domain, rather than 1 2 solely, and is symmetric among multiple BPAs, satisfying our expectation.…”

Section: B the Proposed Belief Jensen-renyi Divergence Measuresupporting

confidence: 72%

Deep Evidential Remote Sensing Landslide Image Classification With a New Divergence, Multiscale Saliency and an Improved Three-Branched Fusion

Zhang,

Cui,

2024

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Hitherto, image-level classification on remote sensing landslide images has been paid attention to, but the accuracy of traditional deep learning-based methods still have room for improvement. The evidence theory is found efficient to boost the accuracy of neural networks, however, the present study argues three challenges that hinder the lead-in of this theory in deep landslide image classification. Aiming at the three problems, this study makes three improvements. For the interpretability and decision-invariance losses of three previous divergences, we propose a Belief Jensen-Renyi divergence with properties proven. To couple the evidence theory with deep remote sensing landslide image classification, a channel-wise multi-scale visual saliency fusion is developed. We additionally find that the channel-wise fusion is capable to reduce false recognition of networks as compared with original RGB images. To avoid decision failures in evidence-theoretic fusion process, we design an interpretability improved three-branched fusion. Experiments on Bijie Landslide dataset corroborate the synergistic benefits of the three improvements, where the proposal is compared with state-ofthe-art image classification backbone networks, remote sensing image scene classifiers, evidence fusion algorithms and versatile evidence-theoretic deep learning classifiers. We also evaluated the new method with two sort of image degradation, as well as an actual scenario in Luding County, China whose data is publicly available. The source code is at https://github.com/defzhangaa/ deeplandslideDS.

show abstract

Section: B the Proposed Belief Jensen-renyi Divergence Measuresupporting

confidence: 72%

Deep Evidential Remote Sensing Landslide Image Classification With a New Divergence, Multiscale Saliency and an Improved Three-Branched Fusion

Zhang,

Cui,

2024

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

show abstract

“…Landslide occurrence is controlled by a variety of conditioning factors, and therefore, reasonable selection of such factors is essential for improving LSM reliability. Drawing on extensive research in the study area (Ma et al, 2020;Wei et al, 2022;Ye et al, 2022), 11 factors were chosen for LSM (Figure 2), i.e., slope, aspect, plan curvature, profile curvature, relief amplitude, annual rainfall, distance to fault, land use, geomorphology, lithology, and distance to river. We divided the original ongoing factors into several subclasses according to their variable impact on landslide occurrence, as shown in Figure 5.…”

Section: Landslide Conditioning Factorsmentioning

confidence: 99%

“…LSM focuses on the quantitative analysis of landslide spatial distribution, using a set of region-specific conditioning factors (Hess et al, 2017). Recently, machine learning (ML) algorithms have shown promising and effective ways of solving non-linear real-world problems with high accuracy and are widely used in LSM, including random forest (RF), support vector machine (SVM), multilayer perceptron (MLP), logistic regression (LGR), decision tree, and artificial neural network (ANN) models (Aditian et al, 2018;Ye et al, 2022). These methods have essential similarities in the way they select critical condition factors, which reduce the impact of highly correlated factors on the generalization ability of the model.…”

Section: Introductionmentioning

confidence: 99%

Generating accurate negative samples for landslide susceptibility mapping: A combined self-organizing-map and one-class SVM method

Tang²,

Wei³

et al. 2023

Front. Earth Sci.

View full text Add to dashboard Cite

The accuracy of data-driven landslide susceptibility mapping (LSM) is closely affected by the quality of non-landslide samples. This research proposes a method combining a self-organizing-map (SOM) and a one-class SVM (SOM-OCSVM) to generate more reasonable non-landslide samples. We designed two steps: first, a random selection, a SOM network, a one class SVM model, and a SOM-OCSVM model were used to generate non-landslide sample datasets. Second, four machine learning models (MLs)—namely logistic regression (LRG), multilayer perceptron (MLP), support vector machine (SVM), and random forest (RF)—were used to verify the effects of four non-landslide sample datasets on LSM. From the region along the Sichuan-Tibet Highway, we selected 11 conditioning factors and 1186 investigated landslides to perform the aforementioned experiments. The results show that the SOM-OCSVM method achieves the highest AUC (>0.94) and minimum standard deviation (<0.081) compared with other methods. Moreover, RF achieves the best performance in different datasets compared with other ML models. The aforementioned results prove that the proposed method can enhance the performance of ML models to produce more reliable LSM.

show abstract

“…Despite the abundance of high-quality tourism resources along these corridors, the plateau's unique natural conditions and underdeveloped infrastructure pose heightened risks to tourists, hindering the development of tourism along these routes [8]. Intriguingly, the southern Sichuan-Tibet line, known as the world's most perilous road due to its complex geological formations and hydrological climates, exemplifies the pronounced safety concerns plaguing tourism corridors in the Tibetan Plateau region [10]. Therefore, urgent scientific assessments of tourism risks are necessary to guide the construction and management of tourism corridors in this region.…”

Section: Introductionmentioning

confidence: 99%

Assessing Risks in Cross-Regional Tourism Corridors: A Case Study of Tibetan Plateau Tourism

Li,

Ye,

2024

IJGI

View full text Add to dashboard Cite

Due to the frequent impact of external risks, scientific tourism risk assessment has become the primary task to be implemented in the process of tourism development. Especially with the development of self-driving travel, cross-regional tourism corridors have become an important tourism carrier. However, compared to traditional fixed-location tourism, cross-regional tourism introduces a more intricate landscape of risks. Therefore, there is a pressing need to assess the tourism risks inherent in these corridors. There are many cross-regional tourism corridors in the Tibetan Plateau, but the natural environment of the Tibetan Plateau brings great risks to these tourism corridors. That is why this study focuses on the Tibetan Plateau’s tourism corridors, employing methodologies such as the Analytic Hierarchy Process, entropy weight method, geographic information systems (GIS) spatial analysis, and others to delve into their tourism risk profiles and the influencing factors. Our findings reveal elevated tourism risks across the Tibetan Plateau’s corridors, notably concentrated along the Yunnan–Tibet Line, north Sichuan–Tibet Line, Xinjiang–Tibet Line, Tangfan Ancient Road, Qinghai–Tibet Line, and south Sichuan–Tibet Line. Furthermore, Geodetector was employed to scrutinize the factors influencing tourism risk within the Tibetan Plateau’s corridors, identifying tourism resource endowment, geographical location, precipitation patterns, and economic foundations as primary influencers. Notably, the interaction between these factors exacerbates the overall tourism risk. These insights significantly contribute to the field of tourism risk research and provide a scientific basis for formulating robust tourism safety management strategies within the Tibetan Plateau region.

show abstract

GIS-based spatial prediction of landslide using road factors and random forest for Sichuan-Tibet Highway

Cited by 22 publications

References 68 publications

Deep Evidential Remote Sensing Landslide Image Classification With a New Divergence, Multiscale Saliency and an Improved Three-Branched Fusion

Deep Evidential Remote Sensing Landslide Image Classification With a New Divergence, Multiscale Saliency and an Improved Three-Branched Fusion

Generating accurate negative samples for landslide susceptibility mapping: A combined self-organizing-map and one-class SVM method

Assessing Risks in Cross-Regional Tourism Corridors: A Case Study of Tibetan Plateau Tourism

Contact Info

Product

Resources

About