CN107358327A - Landslide liability assessment method based on unmanned aerial vehicle remote sensing images - Google Patents
Landslide liability assessment method based on unmanned aerial vehicle remote sensing images Download PDFInfo
- Publication number
- CN107358327A CN107358327A CN201710602102.4A CN201710602102A CN107358327A CN 107358327 A CN107358327 A CN 107358327A CN 201710602102 A CN201710602102 A CN 201710602102A CN 107358327 A CN107358327 A CN 107358327A
- Authority
- CN
- China
- Prior art keywords
- landslide
- grid cell
- liability
- assessment
- influence
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 58
- 238000011156 evaluation Methods 0.000 claims abstract description 64
- 238000013316 zoning Methods 0.000 claims abstract description 15
- 238000012549 training Methods 0.000 claims description 69
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 30
- 238000003325 tomography Methods 0.000 claims description 26
- 239000000284 extract Substances 0.000 claims description 15
- 238000012545 processing Methods 0.000 claims description 9
- 230000011218 segmentation Effects 0.000 claims description 8
- 230000003694 hair properties Effects 0.000 claims description 6
- 238000013507 mapping Methods 0.000 claims description 6
- 238000011002 quantification Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 abstract description 6
- 238000007477 logistic regression Methods 0.000 abstract description 5
- 238000010276 construction Methods 0.000 abstract description 4
- 238000011160 research Methods 0.000 abstract description 4
- 238000013139 quantization Methods 0.000 description 11
- 239000002689 soil Substances 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 238000001514 detection method Methods 0.000 description 3
- 238000010606 normalization Methods 0.000 description 3
- 239000011435 rock Substances 0.000 description 3
- 238000013461 design Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 241000894007 species Species 0.000 description 2
- 239000004575 stone Substances 0.000 description 2
- 241000196324 Embryophyta Species 0.000 description 1
- 235000019738 Limestone Nutrition 0.000 description 1
- 240000004050 Pentaglottis sempervirens Species 0.000 description 1
- 235000004522 Pentaglottis sempervirens Nutrition 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 239000004927 clay Substances 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 239000010459 dolomite Substances 0.000 description 1
- 229910000514 dolomite Inorganic materials 0.000 description 1
- 239000010438 granite Substances 0.000 description 1
- 239000006028 limestone Substances 0.000 description 1
- 239000000843 powder Substances 0.000 description 1
- 238000001556 precipitation Methods 0.000 description 1
- 239000010453 quartz Substances 0.000 description 1
- 239000011044 quartzite Substances 0.000 description 1
- VYPSYNLAJGMNEJ-UHFFFAOYSA-N silicon dioxide Inorganic materials O=[Si]=O VYPSYNLAJGMNEJ-UHFFFAOYSA-N 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012876 topography Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10032—Satellite or aerial image; Remote sensing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30181—Earth observation
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Tourism & Hospitality (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Data Mining & Analysis (AREA)
- Quality & Reliability (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Computational Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Educational Administration (AREA)
- Primary Health Care (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a kind of landslide liability assessment method based on unmanned aerial vehicle remote sensing images,It gathers image as the principal character data of landslide liability evaluation by the use of unmanned plane,With reference to the regional geologic map and meteorological data of evaluation object regional areas,The comprehensive enough refinements of extraction characterize the landslide relevant parameter for treating different regional area landslides correlation properties in domain for assessment as landslide liability factor of influence,And the landslide disaster occurrence Probability Model established using logistic regression method determines to treat the landslide disaster probability of happening of different regional areas in domain for assessment,So that it is determined that treat the landslide liability zoning evaluation result of domain for assessment,Even if being directed to, entire area scope is less to treat domain for assessment,Also liability zoning evaluation of accurately coming down can be subject to,And evaluation result has regional area with strong points,Reliability is high,The characteristics of practicality is good,Can be taken precautions against natural calamities early warning for landslide,Engineering construction detects,It is accurate that the application such as geological research provides,The data reference information of science.
Description
Technical field
The present invention relates to engineering geology and Geologic Hazard Prospecting and electric powder prediction, and in particular to one kind is based on unmanned plane
The landslide liability assessment method of remote sensing image.
Background technology
In recent years, China is due to the Complex Mountain topography and geomorphology considerably beyond the average mountain area area of world land, plus people
The trigger condition such as the aggravation of class engineering activity, extreme rainfall and earthquake, cause the geological disasters such as landslide, avalanche, mud-rock flow extensively with
Educate, the lives and properties to people cause grave danger, and Easy Living Envi-ronment deteriorates.The geological disasters such as pre- landslide-proofing, turn into and ensure mountain area
The significant problem of Easy Living Envi-ronment urgent need to resolve.
The landslide liability assessment method of main flow is only capable of carrying out landslide hazard zoning evaluation in larger region at present, because
Its precision is excessively rough, and the landslide disaster for the regional area that can not be directed in area is easily sent out possibility and evaluated, therefore, it is difficult to
Actual applied to engineering, the actual reference significance for early warning of being taken precautions against natural calamities to mountain area geology landslide is little.
Therefore, how providing can easily send out what possibility was evaluated for the landslide disaster of different regional areas in area
Embodiment, and ensure higher evaluation accuracy, become landslide and take precautions against natural calamities in early warning application technology one and urgently to be resolved hurrily ask
Topic.
The content of the invention
For the deficiencies in the prior art, it is an object of the invention to provide a kind of based on unmanned aerial vehicle remote sensing images
Come down liability assessment method, realizes that easily sending out possibility to the landslide disaster of different regional areas in designated area is evaluated,
And possess higher evaluation accuracy, it is used as the reference information of landslide disaster early warning.
To achieve the above object, present invention employs following technical scheme：
Landslide liability assessment method based on unmanned aerial vehicle remote sensing images, comprises the following steps：
1) regional geologic map, meteorological data and unmanned aerial vehicle remote sensing images for treating domain for assessment are obtained, and treats domain for assessment
Unmanned aerial vehicle remote sensing images carry out data processing, obtain and treat the digital elevation model of domain for assessment；
2) treat domain for assessment and carry out grid division, each grid cell for treating to divide in domain for assessment is considered as a number
According to unit, for treating each grid cell in domain for assessment from the regional geologic map, meteorological data, unmanned plane for treating domain for assessment
The landslide relevant parameter specified is extracted in remote sensing image and digital elevation model respectively, as treating corresponding grid in domain for assessment
Landslide liability factor of influence corresponding to cell；
3) the liability factor of influence that comes down corresponding to each grid cell in domain for assessment is treated respectively to be quantified and returned
One change is handled, it is determined that treating the normalized value of each landslide liability factor of influence corresponding to each grid cell in domain for assessment；
4) the landslide disaster occurrence Probability Model that training obtains is obtained, the landslide disaster occurrence Probability Model is used to indicate
Function corresponding to grid cell respectively between the normalized value of landslide liability factor of influence and landslide disaster probability of happening value reflects
Penetrate relation；It will treat that respectively the normalized value of landslide liability factor of influence substitutes into corresponding to each grid cell in domain for assessment respectively
To landslide disaster occurrence Probability Model, obtain treating each landslide disaster probability of happening value corresponding to grid cell in domain for assessment；
5) according to the landslide disaster probability of happening value for treating each grid cell in domain for assessment, it is determined that treating the cunning of domain for assessment
Slope liability zoning evaluation result.
In the above-mentioned landslide liability assessment method based on unmanned aerial vehicle remote sensing images, further, the landslide disaster occurs
Probabilistic model is obtained and stored by training in advance, is obtained for calling；The method for training landslide disaster occurrence Probability Model
Comprise the following steps：
A) geographic location area of landslide surface unit position is used as instruction known to selection presence in GIS-Geographic Information System
Practice sample areas, obtain regional geologic map, meteorological data and the unmanned aerial vehicle remote sensing images in training sample region, and to training sample
The unmanned aerial vehicle remote sensing images in region carry out data processing, obtain the digital elevation model in training sample region；
B) grid division is carried out to training sample region, it is in place according to the institute of known landslide surface unit in training sample region
Put and determine whether each grid cell divided in training sample region belongs to landslide surface unit cell and marked respectively, and
The each grid cell divided in training sample region is considered as a data cell, for each grid in training sample region
Lattice cell is distinguished from the regional geologic map in training sample region, meteorological data, unmanned aerial vehicle remote sensing images and digital elevation model
Extract the landslide relevant parameter specified, as corresponding to corresponding grid cell in training sample region come down liability influence because
Son；
C) respectively to corresponding to each grid cell in training sample region come down liability factor of influence carry out quantify and
Normalized, determine the respectively normalization of landslide liability factor of influence corresponding to each grid cell in training sample region
Value；
D) several are randomly selected in training sample region labeled as the grid cell of landslide surface unit cell and is not marked
The grid cell of landslide surface unit cell is designated as sample grid cell, by respectively landslide liability influences corresponding to grid cell
The factor establishes the Logic Regression Models of landslide disaster liability as independent variable, will be small as sample grid in training sample region
Respectively the normalized value of landslide liability factor of influence is substituting to patrolling for landslide disaster liability corresponding to each grid cell in area
Collect in regression model, and make landslide disaster probability of happening value corresponding to the grid cell labeled as landslide surface unit cell be 1, make
Unmarked is that landslide disaster probability of happening value is 0 corresponding to the grid cell of landslide surface unit cell, to landslide disaster liability
Logic Regression Models be trained and successive Regression, determine the border intercept in the Logic Regression Models of landslide disaster liability
Correlation coefficient value corresponding to each landslide liability factor of influence of value and grid cell, the correlation coefficient value is referring to
Show the size that the change of the corresponding landslide liability factor of influence value of grid cell influences on landslide disaster probability of happening value changes
Degree, so as to which the Logic Regression Models of the landslide disaster liability to determine correlation coefficient value are used as landslide disaster probability of happening
Model, to indicate the normalized value and landslide disaster probability of happening value of each landslide liability factor of influence corresponding to grid cell
Between Function Mapping relation.
In the above-mentioned landslide liability assessment method based on unmanned aerial vehicle remote sensing images, further, the landslide disaster occurs
Probabilistic model is：
P=1/ (1+e-Y)；
Wherein, P represents landslide disaster probability of happening value, and e is natural constant；Y is the logistic regression of landslide disaster liability
The boundary parameter of model, and：
Y=C0+C1X1+C2X2+...+CNXN；
Wherein, C0For the border values of intercept of the Logic Regression Models of landslide disaster liability, XiRepresent the i-th of grid cell
The normalized value of individual landslide liability factor of influence, CiRepresent corresponding to i-th of landslide liability factor of influence of grid cell
Correlation coefficient value, i ∈ { 1,2 ..., N }, N represent the total number of landslide liability factor of influence corresponding to grid cell.
In the above-mentioned landslide liability assessment method based on unmanned aerial vehicle remote sensing images, preferably, the step
It is rapid d) in, choose as sample grid cell grid cell total quantity be not less than 500, and choose be used as sample grid it is small
The mark in area is the grid cell of unit cell and the quantity ratio of the unmarked grid cell for being landslide surface unit cell
For 1：10~1：1.
In the above-mentioned landslide liability assessment method based on unmanned aerial vehicle remote sensing images, preferably, the step
It is rapid 1) in, the unmanned aerial vehicle remote sensing images of acquisition are to be not less than 1 by the take photo by plane engineer's scale of acquisition of unmanned plane low latitude：2000 height
Clear orthography.
In the above-mentioned landslide liability assessment method based on unmanned aerial vehicle remote sensing images, preferably, the step
It is rapid 2) in, landslide liability factor of influence is extracted as follows corresponding to each grid cell：
Extracted from digital elevation model the elevation of specified location point in grid cell, the gradient, slope position, slope aspect, curvature and
Microrelief；
The lithology of specified location point in grid cell, fold building core portion position are extracted from regional geologic map and away from tomography
Distance；
The rainfall of specified location point in grid cell is extracted from meteorological data；
From unmanned aerial vehicle remote sensing images extract grid cell in specified location point vegetation index, away from water system distance, away from road
Road distance and apart from house distance；
The fold building core portion position of specified location point in grid cell is combined after slope aspect calculates, extract grid cell
The slope type of inclining of middle specified location point；
So as to, with above extracted elevation, the gradient, slope position, slope aspect, curvature, microrelief, away from water system distance, away from road
Distance, apart from house distance, lithology, away from tomography distance, rainfall, vegetation index and slope type of inclining as corresponding to grid cell
Come down liability factor of influence.
In the above-mentioned landslide liability assessment method based on unmanned aerial vehicle remote sensing images, preferably, the step
It is rapid 2) in, landslide liability factor of influence is extracted as follows corresponding to each grid cell：
The elevation of each position point in grid cell, the gradient, slope position, slope aspect, curvature and micro- are extracted from digital elevation model
Landforms, the elevation average of each position point in grid cell, gradient average, slope position average, slope aspect average, curvature average are tried to achieve respectively
And microrelief, and the overall slope of grid cell is determined according to slope position, slope aspect and the microrelief of each position point in grid cell respectively
Position, overall slope aspect and overall microrelief；
Extracted from regional geologic map the lithology of each position point in grid cell, fold building core portion position and away from tomography away from
From, the fold building core portion position average of each position point in grid cell is tried to achieve respectively and away from tomography apart from average, and according to grid
The lithology of each position point determines the overall lithology of grid cell in lattice cell；
The rainfall of each position point in grid cell is extracted from meteorological data, and tries to achieve rainfall average；
From unmanned aerial vehicle remote sensing images extract grid cell in each position point vegetation index, away from water system distance, away from road
Distance and apart from house distance, and try to achieve vegetation index average, away from water system apart from average, away from road distance average and apart from house
Apart from average；
The fold building core portion position of each position point in grid cell is combined after slope aspect calculates, extracted in grid cell
The slope type of inclining of each position point, and determine that the entirety of grid cell is inclined slope type；
So as to above-mentioned resulting elevation average, gradient average, overall slope position, overall slope aspect, curvature average, entirety
Microrelief, away from water system apart from average, away from road distance average, apart from house apart from average, overall lithology, away from tomography distance
Value, rainfall average, vegetation index average and entirety incline slope type as elevation, the gradient, slope position, slope corresponding to grid cell
To, curvature, microrelief, away from water system distance, away from road distance, apart from house distance, lithology, away from tomography distance, rainfall, vegetation
Index and slope type of inclining, and then determine landslide liability factor of influence corresponding to grid cell.
In the above-mentioned landslide liability assessment method based on unmanned aerial vehicle remote sensing images, preferably, the step
It is rapid 3) in, landslide liability factor of influence corresponding to each grid cell quantified as follows and normalization at
Reason：
Influenceed for slope position, slope aspect, microrelief, lithology and this 5 landslide liabilities of slope type of inclining corresponding to grid cell
The factor, quantification treatment is carried out according to presetting quantitative criteria, determines slope position, slope aspect, microrelief, lithology corresponding to grid cell
With the initial value for this 5 landslide liability factors of influence of slope type that incline；For elevation corresponding to grid cell, the gradient, curvature,
Away from water system distance, away from road distance, apart from house distance, away from tomography distance, rainfall and the landslide of vegetation index this 9 liabilities
Factor of influence, then according to respective parameter value as initial value；Then, to elevation corresponding to grid cell, the gradient, slope position,
Slope aspect, curvature, microrelief, away from water system distance, away from road distance, apart from house distance, lithology, away from tomography distance, rainfall, plant
It is normalized by the initial value of index and this 14 landslide liability factors of influence of slope type that incline, obtains grid cell
The normalized value of corresponding 14 landslides liability factor of influence, the mode of normalized are：
Wherein, XiThe normalized value of i-th of landslide liability factor of influence of grid cell is represented,Represent grid cell
I-th of landslide liability factor of influence initial value, i ∈ { 1,2 ..., N }, N represent the easily hair of landslide corresponding to grid cell
The total number of property factor of influence, i.e. N=14；μiI-th of landslide easily hair corresponding to each grid cell in domain for assessment is treated in expression
The average of the initial value of property factor of influence；σiI-th of landslide liability corresponding to each grid cell in domain for assessment is treated in expression
The variance of the initial value of factor of influence.
In the above-mentioned landslide liability assessment method based on unmanned aerial vehicle remote sensing images, preferably, the step
It is rapid 5) in, treated according to the landslide disaster probability of happening value for treating each grid cell in domain for assessment each corresponding in domain for assessment
The concrete mode that the landslide liability of regional area is evaluated is：
Previously according to landslide liability divide from low to high liability is extremely low, liability is low, in liability, liability it is high,
The high five landslides evaluation of probability of occurrence grade of liability, and preset the landslide disaster hair of each landslide evaluation of probability of occurrence grade
Raw probability limit scope；For treating a grid cell in domain for assessment, then occurred according to the landslide disaster of the grid cell
The landslide disaster probability of happening compass for the landslide evaluation of probability of occurrence grade that probable value is fallen into, the grid cell is assessed as
Corresponding landslide evaluation of probability of occurrence grade；Thus, according to the landslide liability for treating each grid cell in domain for assessment of evaluation
Opinion rating, it is determined that treating the landslide liability zoning evaluation result of domain for assessment.
It is preferably, described easy in the above-mentioned landslide liability assessment method based on unmanned aerial vehicle remote sensing images
Hair property is extremely low, liability is low, in liability, liability is high, the landslide calamity of the high five landslides evaluation of probability of occurrence grade of liability
Evil probability of happening compass, be to the span [0,1] of landslide disaster probability of happening value using nature breakpoint method, equal
The segmentation of landslide disaster probability of happening boundary value is carried out every method, Fibonacci method, geometry interval method or manual segmentation method and is determined；Institute
The landslide disaster stated between five landslides evaluation of probability of occurrence grade that manual segmentation method refers to be determined according to priori is sent out
Raw probability limit value, determine the landslide disaster probability of happening compass of five landslides evaluation of probability of occurrence grade.
Compared to prior art, the present invention has the advantages that：
1st, the landslide liability assessment method based on unmanned aerial vehicle remote sensing images of the invention, by the use of unmanned plane gather image as
The principal character data of the liability that comes down evaluation, with reference to the regional geologic map and meteorological data of evaluation object regional areas, comprehensively
The enough refinements of extraction characterize and treat that the landslide relevant parameter of different regional areas landslides correlation properties in domain for assessment is easy as landslide
Hair property factor of influence, and the landslide disaster occurrence Probability Model established using logistic regression method determines to treat difference in domain for assessment
The landslide disaster probability of happening of regional area, so that it is determined that treating the landslide liability zoning evaluation result of domain for assessment.
2nd, the landslide liability assessment method of the invention based on unmanned aerial vehicle remote sensing images, even if being directed to entire area scope
It is less to treat domain for assessment, landslide disaster probability of happening it can be also distributed according to corresponding to treating different grid cells in domain for assessment
Situation, treat domain for assessment and be subject to liability zoning evaluation of accurately coming down, its gained evaluation result has regional area
With strong points, the characteristics of reliability is high, practicality is good, can be landslide take precautions against natural calamities early warning, engineering construction detection, geological research etc. should
With providing accurate, science data reference information.
Brief description of the drawings
Fig. 1 is the flow chart of the landslide liability assessment method of the invention based on unmanned aerial vehicle remote sensing images.
Embodiment
It is easy that accurately landslide disaster can not be carried out in face of existing landslide hazard assessment technology for regional area in area
The problem of hair property evaluation, by analyzing its main cause, it is the assessment parameters that existing landslide hazard assessment technology is utilized
Coverage is larger, and overlay area precision is more rough, can not embody the evaluation between different regional areas in areal
Parameter differences.The characteristics of unmanned air vehicle technique is by its low cost, high security, high maneuverability and high-resolution in recent years, is opened
Beginning is employed for geologic prospect field.Therefore, it may be considered that possess the characteristics of high-resolution using unmanned plane collection image, shoot
The unmanned plane image of domain for assessment is treated, and therefrom extracts to treat to refine in domain for assessment and characterizes the related spy in regional area landslide
Property parameter, treat the evaluation of the landslide disaster liability of regional area in domain for assessment.
Based on above-mentioned technical thought, the invention provides a kind of landslide liability evaluation side based on unmanned aerial vehicle remote sensing images
Method, its flow is as shown in figure 1, comprise the following steps：
1) regional geologic map, meteorological data and unmanned aerial vehicle remote sensing images for treating domain for assessment are obtained, and treats domain for assessment
Unmanned aerial vehicle remote sensing images carry out data processing, obtain and treat the digital elevation model of domain for assessment.
In the step, the unmanned aerial vehicle remote sensing images of acquisition preferably by unmanned plane low latitude take photo by plane acquisition engineer's scale it is not small
In 1：2000 high definition orthography (high definition, i.e. resolution ratio reach more than 720p), preferably to ensure unmanned aerial vehicle remote sensing images
Possess sufficiently high resolution ratio and characterize the landslide correlation properties for treating different regional areas in domain for assessment that can refine.And utilize
Unmanned aerial vehicle remote sensing images handle to obtain the digital elevation model for treating domain for assessment, are to treat the bottom surface landform in domain for assessment
Expression is digitized, different regional area landslides correlation properties in domain for assessment are treated in order to which later stage extraction can refine sign
Landslide relevant parameter.As for handling to obtain the specific Processing Algorithm of respective digital elevation model based on unmanned aerial vehicle remote sensing images,
It has been ripe prior art, and the innovation point of non-invention, therefore it is not deployed to illustrate.
2) treat domain for assessment and carry out grid division, each grid cell for treating to divide in domain for assessment is considered as a number
According to unit, for treating each grid cell in domain for assessment from the regional geologic map, meteorological data, unmanned plane for treating domain for assessment
The landslide relevant parameter specified is extracted in remote sensing image and digital elevation model respectively, as treating corresponding grid in domain for assessment
Landslide liability factor of influence corresponding to cell.
In this step, geographic areas chi corresponding to each grid cell obtained by domain for assessment progress grid division is treated
Very little desirably no more than 100m × 100m is accurate more rationally to treat the landslide liability evaluation of regional area in domain for assessment to guarantee
True property.For example, determining region for county domain, town domain, village domain, different the to be evaluated of area coverage rank in four kinds of place, can draw respectively
It is 100m × 100m, 50m × 50m, 30m × 30m, 10m × 10m grid cell to divide geographic areas size.As for treating area of interest
The grid cell total quantity of domain gained after grid division, then according to treating the gross area of domain for assessment and each grid cell pair
The geographic areas size answered and determined.After treating domain for assessment and carrying out grid division, due to that will treat to be drawn in domain for assessment
The each grid cell divided is considered as a data cell, therefore treats the easily hair of landslide corresponding to each grid cell in domain for assessment
The extraction of property factor of influence, can use different modes of operation.For example, as a kind of embodiment, each grid cell pair
The landslide liability factor of influence answered is extracted as follows：Grid cell middle finger is extracted from digital elevation model
Determine elevation, the gradient, slope position, slope aspect, curvature and the microrelief of location point；Specific bit in grid cell is extracted from regional geologic map
Put lithology a little, fold building core portion position and away from tomography distance；Specified location point in grid cell is extracted from meteorological data
Rainfall；From unmanned aerial vehicle remote sensing images extract grid cell in specified location point vegetation index, away from water system distance, away from road
Road distance and apart from house distance；The fold building core portion position of specified location point in grid cell is combined after slope aspect calculates,
Extract the slope type of inclining of specified location point in grid cell；So as to, with above extracted elevation, the gradient, slope position, slope aspect,
Curvature, microrelief, away from water system distance, refer to away from road distance, apart from house distance, lithology, away from tomography distance, rainfall, vegetation
Number and slope type of inclining are as landslide liability factor of influence corresponding to grid cell.In another example as another embodiment, often
Landslide liability factor of influence is extracted as follows corresponding to individual grid cell：Extracted from digital elevation model
The elevation of each position point, the gradient, slope position, slope aspect, curvature, microrelief, try to achieve each position in grid cell respectively in grid cell
Point elevation average, gradient average, slope position average, slope aspect average, curvature average and microrelief, and according in grid cell everybody
Slope position, slope aspect and the microrelief put a little determine overall slope position, overall slope aspect and the overall microrelief of grid cell respectively；From region
The lithology of each position point in grid cell, fold building core portion position are extracted in geologic map and away from tomography distance, try to achieve grid respectively
In lattice cell the fold building core portion position average of each position point and away from tomography apart from average, and according to each position in grid cell
The lithology of point determines the overall lithology of grid cell；The rainfall of each position point in grid cell is extracted from meteorological data, and
Try to achieve rainfall average；From unmanned aerial vehicle remote sensing images extract grid cell in each position point vegetation index, away from water system distance,
Away from road distance and apart from house distance, and try to achieve vegetation index average, away from water system apart from average, away from road distance average and away from
From average with a distance from house；The fold building core portion position of each position point in grid cell is combined after slope aspect calculates, extract grid
The slope type of inclining of each position point in lattice cell, and determine that the entirety of grid cell is inclined slope type；So as to above-mentioned resulting height
Journey average, gradient average, overall slope position, overall slope aspect, curvature average, overall microrelief, away from water system apart from average, away from road away from
From average, with a distance from house apart from average, overall lithology, away from tomography apart from average, rainfall average, vegetation index average and entirety
Slope type of inclining as elevation corresponding to grid cell, the gradient, slope position, slope aspect, curvature, microrelief, away from water system distance, away from road away from
From, with a distance from house distance, lithology, away from tomography distance, rainfall, vegetation index and slope type of inclining, and then determine grid cell pair
The landslide liability factor of influence answered.14 above-mentioned factors of influence are respectively from geographic and geomorphic conditions, the geology ring for influenceing landslide
Three border condition, induced conditions aspects are chosen after considering and obtained, and can embody landslide liability influence factor comprehensively.
3) the liability factor of influence that comes down corresponding to each grid cell in domain for assessment is treated respectively to be quantified and returned
One change is handled, it is determined that treating the normalized value of each landslide liability factor of influence corresponding to each grid cell in domain for assessment.
It is directed to feelings of the 14 above-mentioned landslide relevant parameters of selection as the landslide liability factor of influence of grid cell
Condition, can be according to such as when for the liability factor of influence that comes down corresponding to each grid cell quantify and normalized
Under type is carried out：
Firstly, for slope position, slope aspect, microrelief, lithology and this 5 landslide liabilities of slope type of inclining corresponding to grid cell
Factor of influence, due to belonging to qualitative parameter rather than quantitative parameter, it is therefore desirable to carried out according to presetting quantitative criteria at quantization
Reason, determine slope position corresponding to grid cell, slope aspect, microrelief, lithology and this 5 landslide liability factors of influence of slope type that incline
Initial value.When it is implemented, this 5 landslide liability factors of influence for belonging to qualitative parameter, can be according to presetting point
Class quantization standard, using classification quantitative label as its respective quantization value mode, carry out the determination of initial value.For example, slope
Position, slope aspect, microrelief, lithology and the respective classification quantitative standard of slope type of inclining and corresponding quantization value mode difference are as follows：
The default classification quantitative standard in slope position and corresponding quantization value mode are：1. mountain valley；2. descending；3. flat slope；4.
Mesoslope；5. go up a slope；6. ridge.
The default classification quantitative standard of slope aspect and corresponding quantization value mode are：1. plane；2. north；3. northeast；4. east；
5. the southeast；6. south；7. southwest；8. west；9. northwest.
The default classification quantitative standard of microrelief and corresponding quantization value mode are：1. valley, deep stream；2. water at mesoslope
System, slack；3. highland water system, water source；4.U types mountain valley；5. Plain；6. spacious slope；7. upslope, tableland；8. local mountain valley
In ridge；9. the ridge at the mesoslope of Plain, hill；10. mountain top, ridge eminence.
The hardness according to slopes lithology of lithology is divided into 7 classes, can according to lithology by firmly to it is soft quantization value by low
To high mode, quantization value is carried out to lithology；Because hardness is not only relevant with Rock And Soil species, also with its rate of decay
It is relevant, cause related lithology species various, here just without enumerating, only choose representative lithology for example, with
Lower Rock And Soil is gentle breeze：1. quartzite, granite；2. quartz porphyry, basalt；3. quartzy sandstone, dolomite；4. stone
Limestone, griotte；5. mud stone, shale；6. clay, gravelly soil；7. fertile soil, bog soil, bird's-eye gravel.
The default classification quantitative standard of slope type of inclining and corresponding quantization value mode are：1. flat folded slope；2. reverse slope/anti-
To slope/introversion slope；3. traversed by slope/tangential slope；4. oblique slope；5. oblique flare slope；6.II types dip slope/flare slope；7.I types
Dip slope/flare slope.
Thus, slope position, slope aspect, microrelief, lithology and this 5 the initial of landslide liability factor of influence of slope type that incline take
Value, can according to its each classification quantitative standard and it is corresponding quantization value mode determined.
Secondly, for elevation corresponding to grid cell, the gradient, curvature, away from water system distance, away from road distance, apart from house
Distance, come down liability factors of influence away from tomography distance, rainfall and vegetation index this 9, then makees according to respective parameter value
For initial value.For example, rainfall can drop in effective precipitation, accumulated rainfall or more annuals according to corresponding to grid cell
The parameter value of rainfall is as initial value.
Then, to elevation corresponding to grid cell, the gradient, slope position, slope aspect, curvature, microrelief, away from water system distance, away from road
Road distance, apart from house distance, lithology, come down liabilities away from tomography distance, rainfall, vegetation index and the slope type this 14 of inclining
The initial value of factor of influence is normalized, and obtains 14 landslides liability factor of influence corresponding to grid cell
Normalized value, the mode of normalized is：
Wherein, XiThe normalized value of i-th of landslide liability factor of influence of grid cell is represented,Represent grid cell
I-th of landslide liability factor of influence initial value, i ∈ { 1,2 ..., N }, N represent the easily hair of landslide corresponding to grid cell
The total number of property factor of influence, i.e. N=14；μiI-th of landslide easily hair corresponding to each grid cell in domain for assessment is treated in expression
The average of the initial value of property factor of influence；σiI-th of landslide liability corresponding to each grid cell in domain for assessment is treated in expression
The variance of the initial value of factor of influence.
4) the landslide disaster occurrence Probability Model that training obtains is obtained, the landslide disaster occurrence Probability Model is used to indicate grid
Function Mapping corresponding to lattice cell between the normalized value of each landslide liability factor of influence and landslide disaster probability of happening value
Relation；It will treat that respectively the normalized value of landslide liability factor of influence is substituting to corresponding to each grid cell in domain for assessment respectively
Landslide disaster occurrence Probability Model, obtain treating each landslide disaster probability of happening value corresponding to grid cell in domain for assessment.
The landslide disaster occurrence Probability Model that application is treated in this step can be obtained and stored by training in advance,
For carrying out acquisition use when needing and calling.
The method of training landslide disaster occurrence Probability Model comprises the following steps：
A) geographic location area of landslide surface unit position is used as instruction known to selection presence in GIS-Geographic Information System
Practice sample areas, obtain regional geologic map, meteorological data and the unmanned aerial vehicle remote sensing images in training sample region, and to training sample
The unmanned aerial vehicle remote sensing images in region carry out data processing, obtain the digital elevation model in training sample region.
In the step, for the training sample region selected by training landslide disaster occurrence Probability Model, theoretically come
Say it can is the geographic location area that landslide surface unit and known landslide surface unit position arbitrarily be present.But in order to
As far as possible improve using training gained landslide disaster occurrence Probability Model treat domain for assessment carry out come down liability evaluation
Accuracy, be preferably selected geographic location area close with the geological environment for treating domain for assessment as training sample region, very
It can extremely use and treat domain for assessment itself as training sample region, as long as treating landslide surface unit itself be present inside domain for assessment
And the position of known landslide surface unit；And each described geological environment type can pass through variant geographic location area
Advance geologic prospect and classification is determined, so as to the geological environment phase known which geographic location area with treat domain for assessment
It is approximate.As in the step be directed to training sample region obtain unmanned aerial vehicle remote sensing images requirement, also with for treating domain for assessment
Unmanned aerial vehicle remote sensing images obtain require identical, preferably by unmanned plane low latitude take photo by plane acquisition engineer's scale be not less than 1：2000
High definition orthography, with preferably ensure unmanned aerial vehicle remote sensing images possess sufficiently high resolution ratio with can refine characterize training
The landslide correlation properties of different regional areas in sample areas.
B) grid division is carried out to training sample region, it is in place according to the institute of known landslide surface unit in training sample region
Put and determine whether each grid cell divided in training sample region belongs to landslide surface unit cell and marked respectively, and
The each grid cell divided in training sample region is considered as a data cell, for each grid in training sample region
Lattice cell is distinguished from the regional geologic map in training sample region, meteorological data, unmanned aerial vehicle remote sensing images and digital elevation model
Extract the landslide relevant parameter specified, as corresponding to corresponding grid cell in training sample region come down liability influence because
Son.
Similarly, geographic areas size corresponding to each grid cell obtained by grid division is carried out to training sample region
It is preferably also and is not more than 100m × 100m, with more rationally to ensures the landslide liability evaluation to regional area in training sample region
Accuracy, equally, the grid cell total quantity of training sample region gained after grid division, then according to training sample region
Geographic areas size corresponding to the gross area and each grid cell and determined.And for each grid in training sample region
Lattice cell extraction as landslide liability factor of influence landslide relevant parameter, it is necessary to for treating the grid in domain for assessment
The landslide liability factor of influence extracting mode of cell is identical, i.e. can all extract elevation, the gradient, slope corresponding to grid cell
Position, slope aspect, curvature, microrelief, away from water system distance, away from road distance, apart from house distance, lithology, away from tomography apart from, rainfall
Amount, vegetation index and this 14 landslide relevant parameters of slope type that incline are as landslide liability factor of influence corresponding to grid cell.
C) respectively to corresponding to each grid cell in training sample region come down liability factor of influence carry out quantify and
Normalized, determine the respectively normalization of landslide liability factor of influence corresponding to each grid cell in training sample region
Value.
Equally, influenceed herein in 14 above-mentioned landslide relevant parameters of selection as the landslide liability of grid cell
The situation of the factor, when for the liability factor of influence that comes down corresponding to each grid cell quantify and normalized,
Also carry out as follows：
Firstly, for slope position, slope aspect, microrelief, lithology and this 5 landslide liabilities of slope type of inclining corresponding to grid cell
Factor of influence determines slope position corresponding to grid cell, slope aspect, micro-, it is necessary to carry out quantification treatment according to presetting quantitative criteria
The initial value of landforms, lithology and this 5 landslide liability factors of influence of slope type that incline.Secondly, for corresponding to grid cell
Elevation, the gradient, curvature, away from water system distance, away from road distance, apart from house distance, away from tomography distance, rainfall and vegetation index
This 9 landslide liability factors of influence, then according to respective parameter value as initial value.Then, to corresponding to grid cell
Elevation, the gradient, slope position, slope aspect, curvature, microrelief, away from water system distance, away from road distance, apart from house distance, lithology, away from break
The come down initial value of liability factors of influence of layer distance, rainfall, vegetation index and the slope type this 14 of inclining is normalized
Processing, obtain the normalized value of 14 landslides liability factor of influence corresponding to grid cell, the mode of normalized
For：
Wherein, XjThe normalized value of j-th of landslide liability factor of influence of grid cell is represented,Represent grid cell
J-th of landslide liability factor of influence initial value, j ∈ { 1,2 ..., N }, N represent the easily hair of landslide corresponding to grid cell
The total number of property factor of influence, i.e. N=14；μjRepresent that j-th of landslide is easy corresponding to each grid cell in training sample region
The average of the initial value of hair property factor of influence；σjRepresent that j-th of landslide is easy corresponding to each grid cell in training sample region
The variance of the initial value of hair property factor of influence.
D) several are randomly selected in training sample region labeled as the grid cell of landslide surface unit cell and is not marked
The grid cell of landslide surface unit cell is designated as sample grid cell, by respectively landslide liability influences corresponding to grid cell
The factor establishes the Logic Regression Models of landslide disaster liability as independent variable, will be small as sample grid in training sample region
Respectively the normalized value of landslide liability factor of influence is substituting to patrolling for landslide disaster liability corresponding to each grid cell in area
Collect in regression model, and make landslide disaster probability of happening value corresponding to the grid cell labeled as landslide surface unit cell be 1, make
Unmarked is that landslide disaster probability of happening value is 0 corresponding to the grid cell of landslide surface unit cell, to landslide disaster liability
Logic Regression Models be trained and successive Regression, determine the border intercept in the Logic Regression Models of landslide disaster liability
Correlation coefficient value corresponding to each landslide liability factor of influence of value and grid cell, the correlation coefficient value is referring to
Show the size that the change of the corresponding landslide liability factor of influence value of grid cell influences on landslide disaster probability of happening value changes
Degree, so as to which the Logic Regression Models of the landslide disaster liability to determine correlation coefficient value are used as landslide disaster probability of happening
Model, to indicate the normalized value and landslide disaster probability of happening value of each landslide liability factor of influence corresponding to grid cell
Between Function Mapping relation.
In the step, the mark that sample grid cell is selected as in training sample region is unit cell
Grid cell and the unmarked grid cell total quantity for landslide surface unit cell are more, then to landslide disaster occurrence Probability Model
Training effect it is better, therefore in training sample region select sample grid cell total quantity be preferably not less than 500,
Even can be more in the case of conditions permit；Meanwhile in order to preferably ensure training precision, selection is used as sample grid cell
Mark be unit cell grid cell and the unmarked grid cell for being landslide surface unit cell quantity ratio most
1 can be reached well：10~1：1.When building the Logic Regression Models of landslide disaster liability, using each corresponding to grid cell
Liability factor of influence come down as independent variable, the boundary parameter side of the Logic Regression Models of landslide disaster liability can be established
Journey：
Y=C0+C1X1+C2X2+...+CNXN；
Wherein, Y is the boundary parameter of the Logic Regression Models of landslide disaster liability；C0For patrolling for landslide disaster liability
Collect the border values of intercept of regression model, XjRepresent the normalized value of i-th of landslide liability factor of influence of grid cell, CjTable
Show correlation coefficient value corresponding to j-th of landslide liability factor of influence of grid cell, j ∈ { 1,2 ..., N }, N represent grid
The total number of landslide liability factor of influence corresponding to lattice cell, the 14 landslide relevant parameters stated in the choice are small as grid
In the case of the landslide liability factor of influence in area, then N=14.And according to Logic Regression Models LogitP=ln [Y/ (1-Y)]
Deformation conversion can obtain landslide disaster probability of happening value P computation model：
P=1/ (1+e-Y)；
Wherein, e is natural constant.
Influenceed in training sample region as the liability that respectively come down corresponding to each grid cell of sample grid cell
The normalized value of the factor is substituted into the boundary parameter equation of above-mentioned Logic Regression Models, and order is labeled as landslide surface unit cell
Grid cell corresponding to landslide disaster probability of happening value be 1, make it is unmarked be landslide surface unit cell grid cell it is corresponding
Landslide disaster probability of happening value be 0, the Logic Regression Models of landslide disaster liability are trained and successive Regression, can
Enough determine each landslide liability of the border values of intercept and grid cell in the Logic Regression Models of landslide disaster liability
Correlation coefficient value corresponding to factor of influence, the correlation coefficient value indicate grid cell corresponding landslide liability influence because
The size degree that the change of subvalue influences on landslide disaster probability of happening value changes.It is determined that border values of intercept C0And each correlation
Property coefficient value C1,C2,...,CNAfterwards, the boundary parameter equation Y=C of Logic Regression Models0+C1X1+C2X2+...+CNXNThen it is able to
It is determined that therefore Logic Regression Models LogitP=ln [Y/ (1-Y)] also just determined, therefore the logistic regression mould can be used
Computation model P=1/ (the 1+e of type deformation conversion-Y) landslide disaster occurrence Probability Model is used as, to indicate that grid cell is corresponding
Each landslide liability factor of influence normalized value and landslide disaster probability of happening value between Function Mapping relation.Thus
It can be seen that for different grid cells, its corresponding landslide disaster being calculated by landslide disaster occurrence Probability Model
For probability of happening value P span in the interval of [0,1], its value is more big, shows that landslide disaster probability of happening is bigger.
After landslide disaster occurrence Probability Model is obtained, then it will treat respectively each corresponding to each grid cell in domain for assessment
The normalized value of landslide liability factor of influence is substituting to landslide disaster occurrence Probability Model, just can be calculated and treat area of interest
Each landslide disaster probability of happening value corresponding to grid cell in domain.
5) according to the landslide disaster probability of happening value for treating each grid cell in domain for assessment, it is determined that treating the cunning of domain for assessment
Slope liability zoning evaluation result.
Obtaining treating in domain for assessment after each landslide disaster probability of happening value corresponding to grid cell, it becomes possible to each
The landslide disaster occurrence risk that the landslide disaster probability of happening value of grid cell is characterized, using existing zoning assessment method,
To determine to treat the landslide liability zoning evaluation result of domain for assessment.When evaluation is embodied, in order that obtaining evaluation result more
Specification, follow-up landslide of being more convenient for are taken precautions against natural calamities and used in the application such as early warning, engineering construction detection, geological research, can be previously according to
Landslide liability divide from low to high liability is extremely low, liability is low, in liability, liability is high, high five cunnings of liability
Slope evaluation of probability of occurrence grade, and the landslide disaster probability of happening compass of each landslide evaluation of probability of occurrence grade is preset,
And liability is extremely low, liability is low, in liability, liability it is high, high this five landslides evaluation of probability of occurrence grade of liability
Landslide disaster probability of happening compass, can be to the span [0,1] of landslide disaster probability of happening value using breaking naturally
Point method, equal intervals method, Fibonacci method, geometry interval method or manual segmentation method carry out landslide disaster probability of happening boundary value minute
Cut and determined, wherein, manual segmentation method refers to the five landslides evaluation of probability of occurrence grade determined according to priori
Between landslide disaster probability of happening boundary value, determine it is described five landslide evaluation of probability of occurrence grade landslide disaster probability of happening
Compass；As for liability determined by any split plot design is extremely low, liability is low, in liability, liability is high, liability
The landslide disaster probability of happening compass of high five landslides evaluation of probability of occurrence grade more tallies with the actual situation, and is directed to difference
The situation of geologic province is not quite similar, and the geologic province situation of practical application can be directed in actual applications, in advance using not
Evaluated after determining this five landslide respective landslide disaster probability of happening compass of evaluation of probability of occurrence grade with split plot design
Contrast and statistics, come judge it is any more tally with the actual situation, and be subject to practical application.Then, for treating in domain for assessment
One grid cell, then the landslide evaluation of probability of occurrence grade fallen into according to the landslide disaster probability of happening value of the grid cell
Landslide disaster probability of happening compass, the grid cell is assessed as the evaluation of probability of occurrence grade that comes down accordingly.And then according to
The landslide evaluation of probability of occurrence grade of each grid cell in domain for assessment is treated in evaluation, it is determined that treating the landslide liability of domain for assessment
Zoning evaluation result；For example, can according to the landslide evaluation of probability of occurrence grade for treating each grid cell in domain for assessment of evaluation,
According to liability is extremely low, liability is low, in liability, liability is high, high this five landslides evaluation of probability of occurrence grade of liability
The mode represented by red turn of green gradual change aberration is respectively adopted, corresponding grid cell is coloured, obtains treating domain for assessment
Come down liability probability distribution grid map, enable to variant grid cell landslide liability probability scenarios pass through it is different
Color is able to more intuitively present, so as to conveniently be determined using the landslide liability probability distribution grid map
Treat the landslide liability zoning evaluation result of domain for assessment.
It can see by above-mentioned flow, the landslide liability assessment method of the invention based on unmanned aerial vehicle remote sensing images, profit
The principal character data evaluated by the use of unmanned plane collection image as landslide liability, with reference to the region for evaluating object regional areas
Matter figure and meteorological data, the comprehensive enough refinements of extraction characterize the landslide for treating different regional area landslides correlation properties in domain for assessment
Relevant parameter is as landslide liability factor of influence, and the landslide disaster occurrence Probability Model established using logistic regression method is true
Surely the landslide disaster probability of happening of different regional areas in domain for assessment is treated, so that it is determined that treating the landslide liability area of domain for assessment
Evaluation result is drawn, entire area scope is less to treat domain for assessment so even being directed to, also being capable of root using the inventive method
According to landslide disaster probability of happening distribution situation corresponding to different grid cells in domain for assessment is treated, treat domain for assessment and be subject to more
Accurately landslide liability zoning evaluation, its gained evaluation result have that regional area is with strong points, reliability is high, practicality is good
The characteristics of, the application such as early warning, engineering construction detection, geological research that can be taken precautions against natural calamities for landslide provides accurate, science data reference
Information.
Finally illustrate, the above embodiments are merely illustrative of the technical solutions of the present invention and it is unrestricted, although with reference to compared with
The present invention is described in detail good embodiment, it will be understood by those within the art that, can be to the skill of the present invention
Art scheme is modified or equivalent substitution, and without departing from the objective and scope of technical solution of the present invention, it all should cover at this
Among the right of invention.
Claims (10)
1. the landslide liability assessment method based on unmanned aerial vehicle remote sensing images, it is characterised in that comprise the following steps：
1) regional geologic map, meteorological data and unmanned aerial vehicle remote sensing images for treating domain for assessment are obtained, and treats the nothing of domain for assessment
Man-machine remote sensing image carries out data processing, obtains the digital elevation model for treating domain for assessment；
2) treat domain for assessment and carry out grid division, each grid cell for treating to divide in domain for assessment is considered as a data sheet
Member, for treating each grid cell in domain for assessment from the regional geologic map, meteorological data, unmanned aerial vehicle remote sensing for treating domain for assessment
The landslide relevant parameter specified is extracted in image and digital elevation model respectively, as treating corresponding grid cell in domain for assessment
Corresponding landslide liability factor of influence；
3) the liability factor of influence that comes down corresponding to each grid cell in domain for assessment is treated respectively to be quantified and normalized
Processing, it is determined that treating the normalized value of each landslide liability factor of influence corresponding to each grid cell in domain for assessment；
4) the landslide disaster occurrence Probability Model that training obtains is obtained, the landslide disaster occurrence Probability Model is used to indicate grid
Function Mapping corresponding to cell respectively between the normalized value of landslide liability factor of influence and landslide disaster probability of happening value is closed
System；It will treat that respectively the normalized value of landslide liability factor of influence is substituting to cunning corresponding to each grid cell in domain for assessment respectively
Slope disaster occurrence Probability Model, obtain treating each landslide disaster probability of happening value corresponding to grid cell in domain for assessment；
5) according to the landslide disaster probability of happening value of each grid cell in domain for assessment is treated, it is determined that treating that the landslide of domain for assessment is easy
Hair property zoning evaluation result.
2. the landslide liability assessment method based on unmanned aerial vehicle remote sensing images according to claim 1, it is characterised in that described
Landslide disaster occurrence Probability Model is obtained and stored by training in advance, is obtained for calling；Landslide disaster is trained to occur general
The method of rate model comprises the following steps：
A) geographic location area of landslide surface unit position is used as training sample known to selection presence in GIS-Geographic Information System
One's respective area, regional geologic map, meteorological data and the unmanned aerial vehicle remote sensing images in training sample region are obtained, and to training sample region
Unmanned aerial vehicle remote sensing images carry out data processing, obtain the digital elevation model in training sample region；
B) grid division is carried out to training sample region, according to the position of known landslide surface unit in training sample region point
Whether each grid cell that Que Ding do not divide in training sample region belongs to landslide surface unit cell and is marked, and will instruction
The each grid cell for practicing division in sample areas is considered as a data cell, small for each grid in training sample region
Extracted respectively from the regional geologic map in training sample region, meteorological data, unmanned aerial vehicle remote sensing images and digital elevation model in area
Go out the landslide relevant parameter specified, as the liability factor of influence that come down corresponding to corresponding grid cell in training sample region；
C) the liability factor of influence that come down corresponding to each grid cell in training sample region is quantified and normalizing respectively
Change is handled, and determines the normalized value of each landslide liability factor of influence corresponding to each grid cell in training sample region；
D) several are randomly selected in training sample region labeled as the grid cell of landslide surface unit cell and unmarked is
The grid cell of landslide surface unit cell is as sample grid cell, by the liability factor of influence that respectively come down corresponding to grid cell
The Logic Regression Models of landslide disaster liability are established as independent variable, using in training sample region as sample grid cell
Respectively the normalized value of landslide liability factor of influence is substituting to the logic time of landslide disaster liability corresponding to each grid cell
Return in model, and order is labeled as landslide disaster probability of happening value corresponding to the grid cell of landslide surface unit cell and is 1, makes and not marking
It is 0 to be designated as landslide disaster probability of happening value corresponding to the grid cell of landslide surface unit cell, and landslide disaster liability is patrolled
Volume regression model is trained and successive Regression, determine border values of intercept in the Logic Regression Models of landslide disaster liability with
And correlation coefficient value corresponding to each landslide liability factor of influence of grid cell, the correlation coefficient value is indicating grid
The size degree that the change of the corresponding landslide liability factor of influence value of lattice cell influences on landslide disaster probability of happening value changes,
So as to the landslide disaster liability to determine correlation coefficient value Logic Regression Models as landslide disaster occurrence Probability Model,
To indicate corresponding to grid cell between the normalized value of each landslide liability factor of influence and landslide disaster probability of happening value
Function Mapping relation.
3. the landslide liability assessment method based on unmanned aerial vehicle remote sensing images according to claim 2, it is characterised in that described
Landslide disaster occurrence Probability Model is：
P=1/ (1+e-Y)；
Wherein, P represents landslide disaster probability of happening value, and e is natural constant；Y is the Logic Regression Models of landslide disaster liability
Boundary parameter, and：
Y=C0+C1X1+C2X2+...+CNXN；
Wherein, C0For the border values of intercept of the Logic Regression Models of landslide disaster liability, XiRepresent i-th of cunning of grid cell
The normalized value of slope liability factor of influence, CiRepresent related corresponding to i-th of landslide liability factor of influence of grid cell
Property coefficient value, i ∈ { 1,2 ..., N }, N represent the total number of landslide liability factor of influence corresponding to grid cell.
4. the landslide liability assessment method based on unmanned aerial vehicle remote sensing images according to claim 2, it is characterised in that described
In step d), choose and be not less than 500 as the grid cell total quantity of sample grid cell, and choose small as sample grid
The mark in area is the grid cell of unit cell and the quantity ratio of the unmarked grid cell for being landslide surface unit cell
For 1：10~1：1.
5. the landslide liability assessment method based on unmanned aerial vehicle remote sensing images according to claim 1, it is characterised in that described
In step 1), the unmanned aerial vehicle remote sensing images of acquisition are to be not less than 1 by the take photo by plane engineer's scale of acquisition of unmanned plane low latitude：2000 height
Clear orthography.
6. the landslide liability assessment method based on unmanned aerial vehicle remote sensing images according to claim 1, it is characterised in that described
In step 2), landslide liability factor of influence is extracted as follows corresponding to each grid cell：
The elevation of specified location point in grid cell, the gradient, slope position, slope aspect, curvature and micro- are extracted from digital elevation model
Looks；
Extracted from regional geologic map the lithology of specified location point in grid cell, fold building core portion position and away from tomography away from
From；
The rainfall of specified location point in grid cell is extracted from meteorological data；
From unmanned aerial vehicle remote sensing images extract grid cell in specified location point vegetation index, away from water system distance, away from road away from
From the house distance with a distance from；
The fold building core portion position of specified location point in grid cell is combined after slope aspect calculates, extract grid cell middle finger
Determine the slope type of inclining of location point；
So as to, with above extracted elevation, the gradient, slope position, slope aspect, curvature, microrelief, away from water system distance, away from road distance,
Apart from house distance, lithology, away from tomography distance, rainfall, vegetation index and slope type of inclining as landslide corresponding to grid cell
Liability factor of influence.
7. the landslide liability assessment method based on unmanned aerial vehicle remote sensing images according to claim 1, it is characterised in that described
In step 2), landslide liability factor of influence is extracted as follows corresponding to each grid cell：
Elevation, the gradient, slope position, slope aspect, curvature and the microrelief of each position point in grid cell are extracted from digital elevation model,
The elevation average of each position point in grid cell, gradient average, slope position average, slope aspect average, curvature average and micro- are tried to achieve respectively
Landforms, and the overall slope position, whole of grid cell is determined according to slope position, slope aspect and the microrelief of each position point in grid cell respectively
Body slope aspect and overall microrelief；
The lithology of each position point in grid cell, fold building core portion position are extracted from regional geologic map and away from tomography distance,
The fold building core portion position average of each position point in grid cell is tried to achieve respectively and away from tomography apart from average, and it is small according to grid
The lithology of each position point determines the overall lithology of grid cell in area；
The rainfall of each position point in grid cell is extracted from meteorological data, and tries to achieve rainfall average；
From unmanned aerial vehicle remote sensing images extract grid cell in each position point vegetation index, away from water system distance, away from road distance
With apart from house distance, and try to achieve vegetation index average, away from water system apart from average, away from road distance average and apart from house distance
Average；
The fold building core portion position of each position point in grid cell is combined after slope aspect calculates, extract in grid cell everybody
Slope type of inclining a little is put, and determines that the entirety of grid cell is inclined slope type；
So as to above-mentioned resulting elevation average, gradient average, overall slope position, overall slope aspect, curvature average, entirety micro-ly
Looks, away from water system apart from average, away from road distance average, apart from house apart from average, overall lithology, away from tomography apart from average, drop
Rainfall average, vegetation index average and entirety incline slope type as elevation, the gradient, slope position, slope aspect, song corresponding to grid cell
Rate, microrelief, away from water system distance, away from road distance, apart from house distance, lithology, away from tomography distance, rainfall, vegetation index
With slope type of inclining, and then landslide liability factor of influence corresponding to grid cell is determined.
8. the landslide liability assessment method based on unmanned aerial vehicle remote sensing images according to claim 6 or 7, it is characterised in that
In the step 3), landslide liability factor of influence corresponding to each grid cell is quantified as follows and normalizing
Change is handled：
For slope position corresponding to grid cell, slope aspect, microrelief, lithology and this 5 landslide liability factors of influence of slope type that incline,
Quantification treatment is carried out according to presetting quantitative criteria, slope position, slope aspect, microrelief, lithology corresponding to grid cell is determined and inclines
The initial value of this 5 landslide liability factors of influence of slope type；For elevation corresponding to grid cell, the gradient, curvature, away from water
System's distance, influence away from road distance, apart from house distance, away from tomography distance, rainfall and the landslide of vegetation index this 9 liabilities
The factor, then according to respective parameter value as initial value；Then, to elevation corresponding to grid cell, the gradient, slope position, slope aspect,
Curvature, microrelief, away from water system distance, refer to away from road distance, apart from house distance, lithology, away from tomography distance, rainfall, vegetation
The initial value of number and this 14 landslide liability factors of influence of slope type that incline is normalized, and it is corresponding to obtain grid cell
14 landslides liability factor of influence normalized value, the mode of normalized is：
<mrow>
<msub>
<mi>X</mi>
<mi>i</mi>
</msub>
<mo>=</mo>
<mfrac>
<mrow>
<msubsup>
<mi>X</mi>
<mi>i</mi>
<mn>0</mn>
</msubsup>
<mo>-</mo>
<msub>
<mi>&mu;</mi>
<mi>i</mi>
</msub>
</mrow>
<msub>
<mi>&sigma;</mi>
<mi>i</mi>
</msub>
</mfrac>
<mo>;</mo>
</mrow>
Wherein, XiThe normalized value of i-th of landslide liability factor of influence of grid cell is represented,Represent the of grid cell
The initial value of i landslide liability factor of influence, i ∈ { 1,2 ..., N }, N represent landslide liability shadow corresponding to grid cell
Ring the total number of the factor, i.e. N=14；μiI-th of landslide liability shadow corresponding to each grid cell in domain for assessment is treated in expression
Ring the average of the initial value of the factor；σiExpression treats that i-th of landslide liability influences corresponding to each grid cell in domain for assessment
The variance of the initial value of the factor.
9. the landslide liability assessment method based on unmanned aerial vehicle remote sensing images according to claim 1, it is characterised in that described
In step 5), treated according to the landslide disaster probability of happening value for treating each grid cell in domain for assessment each corresponding in domain for assessment
The concrete mode that the landslide liability of regional area is evaluated is：
Previously according to landslide liability divide from low to high liability is extremely low, liability is low, in liability, liability is high, Yi Fa
Property high five landslides evaluation of probability of occurrence grades, and the landslide disaster for presetting each landslide evaluation of probability of occurrence grade occur it is general
Rate compass；For treating a grid cell in domain for assessment, then according to the landslide disaster probability of happening of the grid cell
The landslide disaster probability of happening compass for the landslide evaluation of probability of occurrence grade that value is fallen into, the grid cell is assessed as accordingly
Landslide evaluation of probability of occurrence grade；Thus, according to the landslide evaluation of probability of occurrence for treating each grid cell in domain for assessment of evaluation
Grade, it is determined that treating the landslide liability zoning evaluation result of domain for assessment.
10. the landslide liability assessment method based on unmanned aerial vehicle remote sensing images according to claim 8, it is characterised in that institute
State liability is extremely low, liability is low, in liability, liability is high, the cunning of the high five landslides evaluation of probability of occurrence grade of liability
Slope disaster probability of happening compass, it is using nature breakpoint method, phase to the span [0,1] of landslide disaster probability of happening value
Method, Fibonacci method, geometry interval method or manual segmentation method carry out the segmentation of landslide disaster probability of happening boundary value and true at equal intervals
It is fixed；Landslide calamity between the described five evaluation of probability of occurrence grades that come down that the manual segmentation method refers to be determined according to priori
Evil probability of happening boundary value, determine the landslide disaster probability of happening compass of five landslides evaluation of probability of occurrence grade.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710602102.4A CN107358327A (en) | 2017-07-21 | 2017-07-21 | Landslide liability assessment method based on unmanned aerial vehicle remote sensing images |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710602102.4A CN107358327A (en) | 2017-07-21 | 2017-07-21 | Landslide liability assessment method based on unmanned aerial vehicle remote sensing images |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107358327A true CN107358327A (en) | 2017-11-17 |
Family
ID=60285277
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710602102.4A Pending CN107358327A (en) | 2017-07-21 | 2017-07-21 | Landslide liability assessment method based on unmanned aerial vehicle remote sensing images |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107358327A (en) |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107957982A (en) * | 2017-12-05 | 2018-04-24 | 中国科学院遥感与数字地球研究所 | Secondary Geological Hazards liability fast evaluation method and system after shake |
| CN108320461A (en) * | 2018-02-08 | 2018-07-24 | 中国地质大学（武汉） | A kind of landslide disaster monitoring device and its monitoring method |
| CN109165424A (en) * | 2018-08-03 | 2019-01-08 | 四川理工学院 | A kind of landslide assessment of easy generation method based on domestic GF-1 satellite data |
| CN109214643A (en) * | 2018-07-18 | 2019-01-15 | 重庆大学 | A kind of Mountainous Building toughness evaluation method based on big data |
| CN109584510A (en) * | 2018-11-30 | 2019-04-05 | 中国公路工程咨询集团有限公司 | A kind of road landslide of high slope disaster alarm method based on valuation functions training |
| CN110427857A (en) * | 2019-07-26 | 2019-11-08 | 国网湖北省电力有限公司检修公司 | A kind of transmission line of electricity geological disasters analysis method based on Remote Sensing Data Fusion Algorithm |
| CN111339826A (en) * | 2020-05-06 | 2020-06-26 | 山西大学 | Landslide unmanned aerial vehicle linear sensor network frame detection system |
| CN111539450A (en) * | 2020-03-24 | 2020-08-14 | 中南大学 | Landslide incidence evaluation method for seismic region |
| CN111582615A (en) * | 2019-02-18 | 2020-08-25 | 中国科学院深圳先进技术研究院 | Evaluation method and system based on LOGISTIC regression model |
| CN111858803A (en) * | 2020-07-06 | 2020-10-30 | 东华理工大学 | Landslide land disaster risk zoning map generation method |
| CN111967544A (en) * | 2020-10-23 | 2020-11-20 | 成都信息工程大学 | Soil landslide disaster space-time intelligent pre-warning method and system |
| CN112232526A (en) * | 2020-09-28 | 2021-01-15 | 中山大学 | Geological disaster susceptibility evaluation method and system based on integration strategy |
| CN113034499A (en) * | 2021-05-24 | 2021-06-25 | 西南交通大学 | Slope surface danger area prediction method, device, equipment and readable storage medium |
| CN113240353A (en) * | 2021-07-12 | 2021-08-10 | 南京擎天全税通信息科技有限公司 | Cross-border e-commerce oriented export factory classification method and device |
| CN113343563A (en) * | 2021-05-27 | 2021-09-03 | 中交第二公路勘察设计研究院有限公司 | Landslide susceptibility evaluation method based on automatic sample selection and surface deformation rate |
| CN113449469A (en) * | 2021-06-22 | 2021-09-28 | 中国地质调查局武汉地质调查中心 | Multi-source information fusion-based slope shallow surface geotechnical engineering property zoning method |
| CN114036841A (en) * | 2021-11-10 | 2022-02-11 | 云南大学 | Landslide incidence prediction method and system based on semi-supervised support vector machine model |
| CN114265431A (en) * | 2021-12-22 | 2022-04-01 | 中国地质大学（武汉） | Ground disaster monitoring method and system based on distributed coverage control |
| CN116151483A (en) * | 2023-04-14 | 2023-05-23 | 山东理工大学 | Regional rainfall landslide probabilistic prediction method and prediction terminal |
| CN118135744A (en) * | 2024-04-30 | 2024-06-04 | 泉州中科星桥空天技术有限公司 | Geological disaster rapid assessment method based on remote sensing technology |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20100113404A (en) * | 2009-04-13 | 2010-10-21 | 주식회사 삼안 | Prediction method of slope failure to real-time in revetment facilities |
| CN104299365A (en) * | 2014-08-06 | 2015-01-21 | 江苏恒创软件有限公司 | Method for monitoring mountain landslide and debris flow in mountainous areas based on unmanned aerial vehicle |
-
2017
- 2017-07-21 CN CN201710602102.4A patent/CN107358327A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20100113404A (en) * | 2009-04-13 | 2010-10-21 | 주식회사 삼안 | Prediction method of slope failure to real-time in revetment facilities |
| CN104299365A (en) * | 2014-08-06 | 2015-01-21 | 江苏恒创软件有限公司 | Method for monitoring mountain landslide and debris flow in mountainous areas based on unmanned aerial vehicle |
Non-Patent Citations (2)
| Title |
|---|
| OZGUN AKCAY: "Landslide Fissure Inference Assessment by ANFIS and Logistic Regression Using UAS-Based Photogrammetry", 《ISPRS INTERNATIONAL JOURNAL OF GEO-INFORMATION》 * |
| 胡东萍: "基于地理空间数据的地震诱发滑坡易感性统计模型研究", 《中国优秀硕士学位论文全文数据库基础科学辑》 * |
Cited By (31)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107957982A (en) * | 2017-12-05 | 2018-04-24 | 中国科学院遥感与数字地球研究所 | Secondary Geological Hazards liability fast evaluation method and system after shake |
| CN107957982B (en) * | 2017-12-05 | 2020-11-20 | 中国科学院遥感与数字地球研究所 | Method and system for rapidly evaluating easiness of post-earthquake secondary geological disasters |
| CN108320461A (en) * | 2018-02-08 | 2018-07-24 | 中国地质大学（武汉） | A kind of landslide disaster monitoring device and its monitoring method |
| CN109214643B (en) * | 2018-07-18 | 2021-10-08 | 重庆大学 | Mountain land building toughness evaluation method based on big data |
| CN109214643A (en) * | 2018-07-18 | 2019-01-15 | 重庆大学 | A kind of Mountainous Building toughness evaluation method based on big data |
| CN109165424A (en) * | 2018-08-03 | 2019-01-08 | 四川理工学院 | A kind of landslide assessment of easy generation method based on domestic GF-1 satellite data |
| CN109584510A (en) * | 2018-11-30 | 2019-04-05 | 中国公路工程咨询集团有限公司 | A kind of road landslide of high slope disaster alarm method based on valuation functions training |
| CN109584510B (en) * | 2018-11-30 | 2021-02-02 | 中国公路工程咨询集团有限公司 | Road high slope landslide hazard early warning method based on evaluation function training |
| CN111582615A (en) * | 2019-02-18 | 2020-08-25 | 中国科学院深圳先进技术研究院 | Evaluation method and system based on LOGISTIC regression model |
| CN110427857A (en) * | 2019-07-26 | 2019-11-08 | 国网湖北省电力有限公司检修公司 | A kind of transmission line of electricity geological disasters analysis method based on Remote Sensing Data Fusion Algorithm |
| CN111539450A (en) * | 2020-03-24 | 2020-08-14 | 中南大学 | Landslide incidence evaluation method for seismic region |
| CN111539450B (en) * | 2020-03-24 | 2023-10-24 | 中南大学 | Landslide susceptibility evaluation method in earthquake region |
| CN111339826A (en) * | 2020-05-06 | 2020-06-26 | 山西大学 | Landslide unmanned aerial vehicle linear sensor network frame detection system |
| CN111339826B (en) * | 2020-05-06 | 2023-05-02 | 山西大学 | Landslide unmanned aerial vehicle linear sensor network frame detecting system |
| CN111858803A (en) * | 2020-07-06 | 2020-10-30 | 东华理工大学 | Landslide land disaster risk zoning map generation method |
| CN111858803B (en) * | 2020-07-06 | 2021-09-21 | 东华理工大学 | Landslide land disaster risk zoning map generation method |
| CN112232526A (en) * | 2020-09-28 | 2021-01-15 | 中山大学 | Geological disaster susceptibility evaluation method and system based on integration strategy |
| CN112232526B (en) * | 2020-09-28 | 2023-10-13 | 中山大学 | Geological disaster vulnerability evaluation method and system based on integration strategy |
| CN111967544A (en) * | 2020-10-23 | 2020-11-20 | 成都信息工程大学 | Soil landslide disaster space-time intelligent pre-warning method and system |
| CN113034499A (en) * | 2021-05-24 | 2021-06-25 | 西南交通大学 | Slope surface danger area prediction method, device, equipment and readable storage medium |
| CN113034499B (en) * | 2021-05-24 | 2021-09-21 | 西南交通大学 | Slope surface danger area prediction method, device, equipment and readable storage medium |
| CN113343563B (en) * | 2021-05-27 | 2022-05-03 | 中交第二公路勘察设计研究院有限公司 | Landslide susceptibility evaluation method based on automatic sample selection and surface deformation rate |
| CN113343563A (en) * | 2021-05-27 | 2021-09-03 | 中交第二公路勘察设计研究院有限公司 | Landslide susceptibility evaluation method based on automatic sample selection and surface deformation rate |
| CN113449469A (en) * | 2021-06-22 | 2021-09-28 | 中国地质调查局武汉地质调查中心 | Multi-source information fusion-based slope shallow surface geotechnical engineering property zoning method |
| CN113240353B (en) * | 2021-07-12 | 2021-11-02 | 南京擎天全税通信息科技有限公司 | Cross-border e-commerce oriented export factory classification method and device |
| CN113240353A (en) * | 2021-07-12 | 2021-08-10 | 南京擎天全税通信息科技有限公司 | Cross-border e-commerce oriented export factory classification method and device |
| CN114036841A (en) * | 2021-11-10 | 2022-02-11 | 云南大学 | Landslide incidence prediction method and system based on semi-supervised support vector machine model |
| CN114265431A (en) * | 2021-12-22 | 2022-04-01 | 中国地质大学（武汉） | Ground disaster monitoring method and system based on distributed coverage control |
| CN116151483A (en) * | 2023-04-14 | 2023-05-23 | 山东理工大学 | Regional rainfall landslide probabilistic prediction method and prediction terminal |
| CN118135744A (en) * | 2024-04-30 | 2024-06-04 | 泉州中科星桥空天技术有限公司 | Geological disaster rapid assessment method based on remote sensing technology |
| CN118135744B (en) * | 2024-04-30 | 2024-08-20 | 泉州中科星桥空天技术有限公司 | Geological disaster rapid assessment method based on remote sensing technology |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107358327A (en) | Landslide liability assessment method based on unmanned aerial vehicle remote sensing images | |
| Veh et al. | Detecting Himalayan glacial lake outburst floods from Landsat time series | |
| CN111858803B (en) | Landslide land disaster risk zoning map generation method | |
| CN109165424A (en) | A kind of landslide assessment of easy generation method based on domestic GF-1 satellite data | |
| Kheir et al. | Spatial soil zinc content distribution from terrain parameters: A GIS-based decision-tree model in Lebanon | |
| Quintus et al. | The efficacy and analytical importance of manual feature extraction using lidar datasets | |
| Guerra et al. | Soil erosion and conservation in Brazil | |
| CN106485254A (en) | A kind of electric transmission line channel topographical features change extraction method | |
| Bou Kheir et al. | The application of GIS based decision-tree models for generating the spatial distribution of hydromorphic organic landscapes in relation to digital terrain data | |
| Langlois et al. | Landscape analysis of nutrient-enriched margins (lagg) in ombrotrophic peatlands | |
| CN114881457B (en) | Landslide and collapse disaster classification method based on decision tree and electronic equipment | |
| Patle et al. | Application of geoinformatics with frequency ratio (FR) model to delineate different groundwater potential zones in Ken Basin, India | |
| Oller et al. | Characterizing major avalanche episodes in space and time in the twentieth and early twenty-first centuries in the Catalan Pyrenees | |
| CN113516060A (en) | Building space fusion discrimination method and device based on multi-temporal high-resolution data | |
| CN110111376B (en) | Coal mining subsidence water area calculation method | |
| Wang et al. | Quantitative relationship between urban green canopy area and urban greening land area | |
| Queen | Large-scale mapping and geomorphometry of upland periglacial landscapes in Eastern Beringia | |
| Badawy | Applications of remote sensing and geographic information systems in geomorphological studies: Safaga-El Quseir area, Red Sea coast, Egypt as an example | |
| D'Addario | A New Approach to Assess the Susceptibility to Shallow Landslides at Regional Scale as Influenced by Bedrock Geo-Mechanical Properties | |
| CN110287915A (en) | A kind of urban impervious surface extracting method based on Landsat remote sensing image | |
| Omar et al. | Production of High-Resolution Map of Biomass Carbon over Forests in Malaysia Estimated using Aboveground Carbon Density Indicator (ACDI) and a Collection of 12 Years Inventory Data | |
| CN114118848B (en) | Quantitative recognition and disaster risk assessment method for urban rock cavernous factors | |
| Koja | Sand dune movement and its impact on human activities in the North Western coast region of Libya. An analysis of the sediment characteristics of sand dunes, and their movement using satellite images, and the effects of encroachment on farms assessed by a questionnaire survey. | |
| Abolfathi et al. | Evaluation of Land Cover Changes Ysing Remote Sensing Technique (Case study: Hableh Rood Subwatershed of Shahrabad Basin) | |
| Cheng et al. | Random Forests for Landslide Prediction in Tsengwen River Watershed, Central Taiwan. Remote Sens. 2021, 13, 199 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171117 |
|
| RJ01 | Rejection of invention patent application after publication |