- RIKEN Center for Advanced Intelligence Project, Goal-Oriented Technology Research Group, Disaster Resilience Science Team - Individual Member of the International Society for Photogrammetry and Remote Sensing Conditioning Factors Determination for Landslide Susceptibility Mapping Using Support Vector Machine Learning
18
Embed
Conditioning Factors Determination for Landslide ...
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
- RIKEN Center for Advanced Intelligence Project, Goal-Oriented Technology Research Group, Disaster Resilience Science Team
- Individual Member of the International Society for Photogrammetry and Remote Sensing
Conditioning Factors Determination for Landslide Susceptibility
Mapping Using Support Vector Machine Learning
Landslides cause
high fatality rates
huge property losses
Landslide analysis can help
to detect areas prone tolandslides,
to provide early warningfor affected residents.
Landslide analysis
Landslide initiation
Landslide Susceptibility
Risk assessments
Landslide Susceptibility
specifically looking at the contribution ofindividual conditioning variables (orfactors).
hydrological and
conditioning variables
geomorphical,
topographical
Identifying the appropriate conditioningfactors is important specially whenconstructing a model to predict potentiallandslide area.
This research seeks to expand on previous works, and answer the following questions:
(1) Despite the existing pool of landslide factors, which of these factors best predict landslides susceptibility?
(2) What is the minimum number of factors to construct a model to come up with a consistent landslide potential map?
(1) Slope angle (8) Stream Power Index (SPI)
(2) Slope aspect (9) Topographic Roughness Index (TRI)
(3) Elevations (10) Sediment TransportIndex (STI)
(4) Total curvature (11) Landuse-Landcover(5) Profile curvature (12) Geology(6) Plan curvature (13) Distance from rivers(7) Topographic Wetness Index (TWI)
(14) Distance to fault
To determine whether or not adding selected factors will improve the prediction of landslide susceptibility.
To evaluate the performance of the SVM model based on the selected group.
Basically, the SVM tries to discover an optimalseparating hyperplane that could effectivelyseparate the input features of two classes withmaximum margin.
α»π¦π(π€ β π₯π + π
β₯ 1β πΏπ
π° is the coefficient vector that defines the hyperplane orientation in the feature
space.
π is the offset of the hyperplane from the origin and
πΉπ the positive slack variables
variance-inflated factor (VIF)
ππΌπΉ =1
1 β π β²2
where π β² represent the multi correlation coefficient between individual
feature and the other features in the model.
In the current study, factors with a ππΌπΉ greater than 5 or 10 were
identified as the high correlation and should be removed.
Pearson's correlation coefficients method
ππ₯π¦ = Οπ=1π ππβ ΰ΄€π
Οπ=1π (ππβ ΰ΄€πα»2
ππβΰ΄€π
Οπ=1π (ππβΰ΄€πα»
2
where ππand ππ are the values of π and π for the πth individual.
A high level of colinearity is identified when the Pearsonβs correlation coefficient is greater than 0.7.
Cohenβs kappa index
πΎ =ππππ βπππ₯π
1βπππ₯π
ππππ denotes the correctly classified proportion of landslideand non-landslide pixels.
πππ₯π indicates the proportion of pixels expected to show
agreement, on the basis of chance.
The area under the receiver operatingcharacteristic curve (AUC) by evaluation theprediction and success rates was looked at toevaluate the performance of both SVMs.
Values from
0.5-0.6 indicates poor,
0.6-0.7 average
0.7-0.8 as good
0.8-0.9 means very good
0.9-1 is exceptional (or excellent)
Training points% Testing points%G1 68% 74%G2 80% 81%
ACCURACY OF THE SVM MODEL FOR BOTH G1 AND G2
DATASETS.
NoConditioning factors
VIF
1 Aspect 1.0119662 TWI 1.333633 TRI 9.3157514 SPI 7.6772495 STI 8.5552346 Geology 1.0700037 Landuse 1.0244538 Plan Curvature 4.33E+139 Profile Curvature 9.01E+1310 Total Curvature 1.88E+1411 Slope 7.02952112 Distance to Fault 1.01305413 Distance to River 1.01205414 Altitude 3.521458
The Estimated Variance Information Factor (VIF) for