CN104239885B - A kind of earthquake disaster damage degree appraisal procedure based on unmanned plane - Google Patents

A kind of earthquake disaster damage degree appraisal procedure based on unmanned plane Download PDF

Info

Publication number
CN104239885B
CN104239885B CN201410454210.8A CN201410454210A CN104239885B CN 104239885 B CN104239885 B CN 104239885B CN 201410454210 A CN201410454210 A CN 201410454210A CN 104239885 B CN104239885 B CN 104239885B
Authority
CN
China
Prior art keywords
image
mrow
region
msub
earthquake disaster
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.)
Active
Application number
CN201410454210.8A
Other languages
Chinese (zh)
Other versions
CN104239885A (en
Inventor
刘皓挺
王巍
王学锋
于文鹏
王军龙
蓝天
马建立
何哲玺
宋伟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
China Aerospace Times Electronics Corp
Original Assignee
China Aerospace Times Electronics Corp
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by China Aerospace Times Electronics Corp filed Critical China Aerospace Times Electronics Corp
Priority to CN201410454210.8A priority Critical patent/CN104239885B/en
Publication of CN104239885A publication Critical patent/CN104239885A/en
Application granted granted Critical
Publication of CN104239885B publication Critical patent/CN104239885B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K9/00Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
    • G06K9/00624Recognising scenes, i.e. recognition of a whole field of perception; recognising scene-specific objects
    • G06K9/0063Recognising patterns in remote scenes, e.g. aerial images, vegetation versus urban areas

Abstract

A kind of earthquake disaster damage degree appraisal procedure based on unmanned plane, hovered using unmanned plane in earthquake disaster area overhead and IMAQ, house is directed to respectively, road, bridge, dykes and dams, massif, river course and the quasi-representative atural object of vegetation seven, the circle that disaster area interesting image regions are carried out according to man-machine interactively formula method selects, and selectively calculate the gray level co-occurrence matrixes textural characteristics of selected areas, Tamura textural characteristics, Gabor wavelet textural characteristics and straight line and circle feature, the final assessment that earthquake disaster atural object damage degree is carried out using support vector machine classifier.This method is more suitable for carrying out information and the assessment of early stage in earthquake and the secondary disaster generation area as caused by earthquake.

Description

A kind of earthquake disaster damage degree appraisal procedure based on unmanned plane
Technical field
The present invention relates to a kind of earthquake disaster damage degree appraisal procedure based on unmanned plane.
Background technology
China's area is vast in territory, and mountain range rises and falls, and in length and breadth, geological conditions is complex in river.Due to climatic effect, The reasons such as geological conditions difference, many areas are easily encroached on by earthquake and its secondary disaster throughout the year.As the Qinghai in China, Yunnan, The area such as western Sichuan, SOUTH OF GANSU is located in the place that Alps-Himalaya earthquake zone is passed through, and repeatedly occurs in recent years More than 4.5 grades of earthquake is crossed, causes larger casualties and economic loss.Under normal circumstances, when earthquake disaster occurs, most Conventional mode is the assessment of earthquake magnitude and degree of splitting index progress disaster loss grade that different earthquake sampled point is collected using seismic detector, this The advantages of kind method is that measurement result is authoritative, versatile;But shortcoming is that measurement result is more abstract, fuzzy or even inaccurate, The degree that is damaged in a clear specific disaster area can not be described.If the definition of traditional earthquake destructiveness is according to the big of earthquake magnitude It is small be divided into weak shock (earthquake magnitude be less than 3 grades), felt earthquake (earthquake magnitude is more than or equal to 3 grades, less than 4.5 grades), (earthquake magnitude is more than middle macroseism 4.5 grades, less than 6 grades), macroseism (earthquake magnitude be more than 6 grades), Giant Bullous (earthquake magnitude is more than 8 grades) etc.;This kind of evaluation index is carrying out ground More obscured for the information that the rescue personnel of non-geology specialty provides when shaking disaster assistance, can not judge that earthquake is built to earth's surface Build and structure caused by damaged condition, and due to seismic detector sampled point limited amount, have necessarily away from different earthquake disaster area Distance, geology characteristic, the construction quality of each department differ, and is actually destroyed caused by earthquake and to injure situation also incomplete It is corresponding with These parameters, therefore be difficult often the one specific real disaster-stricken situation information in earthquake areas of reflection, and then also very Hardly possible judges the emergency degree of seismic disaster relief.
With the development of scientific and technological level, in Disaster Assessment technological means, disaster feelings are carried out using remote sensing or Aerial Images The information of condition has the characteristics of data are directly perceived, credible with assessing, in recent years domestic all kinds of disaster alarms and rescue task It is widely used.Chinese patent " remote sensing earthquake damage information extraction and method for digging based on pixel " (domestic patent publication No. CN101788685A a kind of method using remote Sensing Image Analysis) is proposed, is made a variation by spectrum, principal component fusion is found automatically Region of variation, then region of variation is split using dual threshold algorithm of region growing, the side explained finally by artificial visual Formula extracts Earthquake damage information;Chinese patent " a kind of disaster situation obtaining system " (domestic patent publication No. CN201672918U) proposes to use Unmanned plane carry video camera, digital camera, the method for forward looking infrared image collecting device, gather disaster-stricken image, to aid in calamity Evil rescue and post-disaster reconstruction;A kind of Chinese patent " disaster monitoring electronics delineation system based on unmanned plane " (domestic patent Publication number CN102419171A) propose by the way of the information that a kind of ground delineation system is passed back to unmanned plane uses writing pencil It is labeled.However, it is not difficult to find out by the analysis to above-mentioned patent:The problem of disaster loss grade assessment being carried out using remotely-sensed data Be remotely-sensed data acquisition and editor analytical cycle is longer, cost is higher, the issue and acquisition of data can only be by specific Channel and authoritative department provide;And the acquisition of high-altitude Aerial Images needs to perform flight times using medium-sized or large-scale aircraft Business, is not easy to the carrying and use of rescue group.
The content of the invention
The technical problems to be solved by the invention are to propose a kind of low earthquake disaster damage based on unmanned plane of cost Degree of ruining appraisal procedure, the quick collection to earthquake-stricken area disaster-stricken situation can be realized with assessing.
The present invention includes following technical scheme:
1st, a kind of earthquake disaster damage degree appraisal procedure based on unmanned plane, it is characterised in that realize that step is as follows:
(1) typical earthquake disaster image feature data collection is established, the typical earthquake disaster image feature data collection includes House, road, bridge, dykes and dams, massif, river course and the earthquake disaster of vegetation image feature data collection;
To the multiple earthquake disaster images obtained based on unmanned plane in the past, determined by way of user's hand labeled Go out the image-region of the image-region acquisition more than 200 houses in house, the image-region in each house to determining calculates The characteristics of image in house, and damage degree in house corresponding to the image-region in each house is provided, so as to obtain the earthquake disaster in house Evil image feature data collection;
To the multiple earthquake disaster images obtained based on unmanned plane in the past, determined by way of user's hand labeled Go out image-region of the image-region acquisition more than 200 roads of road, the image-region of each road to determining calculates The characteristics of image of road, and road damage degree corresponding to the image-region of each road is provided, so as to obtain the earthquake disaster of road Evil image feature data collection;
To the multiple earthquake disaster images obtained based on unmanned plane in the past, determined by way of user's hand labeled Go out image-region of the image-region acquisition more than 200 bridges of bridge, the figure of bridge is calculated to the image-region of each bridge As feature, and bridge damage degree corresponding to the image-region of each bridge is provided, so as to obtain the earthquake disaster image of bridge spy Levy data set;
To the multiple earthquake disaster images obtained based on unmanned plane in the past, determined by way of user's hand labeled Go out image-region of the image-region acquisition more than 200 dykes and dams of dykes and dams, the figure of dykes and dams is calculated to the image-region of each dykes and dams As feature, and dykes and dams damage degree corresponding to the image-region of each dykes and dams is provided, so as to obtain the earthquake disaster image of dykes and dams spy Levy data set;
To the multiple earthquake disaster images obtained based on unmanned plane in the past, determined by way of user's hand labeled Go out image-region of the image-region acquisition more than 200 massifs of massif, the figure of massif is calculated to the image-region of each massif As feature, and massif damage degree corresponding to the image-region of each massif is provided, so as to obtain the earthquake disaster image of massif spy Levy data set;
To the multiple earthquake disaster images obtained based on unmanned plane in the past, determined by way of user's hand labeled Go out the image-region of the image-region acquisition more than 200 river courses in river course, the figure in river course is calculated to the image-region in each river course As feature, and damage degree in river course corresponding to the image-region in each river course is provided, so as to obtain the earthquake disaster image in river course spy Levy data set;
To the multiple earthquake disaster images obtained based on unmanned plane in the past, determined by way of user's hand labeled Go out image-region of the image-region acquisition more than 200 vegetation of vegetation, the figure of vegetation is calculated to the image-region of each vegetation As feature, and vegetation damage degree corresponding to the image-region of each vegetation is provided, so as to obtain the earthquake disaster image of vegetation spy Levy data set;
The characteristics of image in house includes gray level co-occurrence matrixes angular second moment index GLCMASM, gray level co-occurrence matrixes index of correlation GLCMCOR, gray level co-occurrence matrixes entropy index GLCMENT, contrast index T in Tamura textural characteristicscon, Gabor direction characters Index fgabor, utilize Hough transform carry out straight-line detection straight line number index HlLoop truss is carried out with using Hough transform Circle number index Hc
Road, bridge, the characteristics of image in dykes and dams and river course include gray level co-occurrence matrixes angular second moment index GLCMASM, ash Spend co-occurrence matrix the moment of inertia index GLCMCON, gray level co-occurrence matrixes entropy index GLCMENT, gray level co-occurrence matrixes inverse difference moment index GLCMIDM, contrast index T in Tamura textural characteristicscon, Gabor direction character indexs fgaborEnter with using Hough transform The straight line number index H of row straight-line detectionl
Massif and the characteristics of image of vegetation include gray level co-occurrence matrixes angular second moment index GLCMASM, gray level co-occurrence matrixes The moment of inertia index GLCMCON, gray level co-occurrence matrixes inverse difference moment index GLCMIDM, contrast index T in Tamura textural characteristicscon With Gabor direction character indexs fgabor
(2) house, road, bridge, dykes and dams, massif, river course and the training of the earthquake disaster of vegetation image feature data are utilized Practice respective support vector machine classifier;
(3) new earthquake disaster image is obtained based on unmanned plane;
(4) house, road, bridge are directed to respectively by way of user's hand labeled to the new earthquake disaster image of acquisition Beam, dykes and dams, massif, river course, the quasi-representative atural object of vegetation 7 determine corresponding image-region, according to the image-region meter determined Characteristics of image corresponding to every quasi-representative atural object is calculated, the last characteristics of image according to corresponding to per quasi-representative atural object uses corresponding support The grader of vector machine carries out the assessment of all kinds of atural object damage degree.
The present invention has the beneficial effect that compared with prior art:
(1) present invention devises a kind of earthquake disaster damage degree appraisal procedure using unmanned plane shooting image, direct root The damage degree that earthquake and its secondary disaster are carried out according to the destroyed degree of earth's surface culture and Images of Natural Scenery is assessed, and is had The characteristics of accurate, directly perceived, is described to disaster area disaster-stricken situation.
(2) the damage degree appraisal procedure designed by the present invention, model is carried out using the image gathered in conventional earthquake disaster The extraction and calculating of feature, accumulation of knowledge is realized with reusing, there is higher assessment reliability and the degree of accuracy.
(3) the disaster damage degree appraisal procedure designed by the present invention, view data is carried out using unmanned aerial vehicle platform near the ground Collection, and the processing that software carries out view data is assessed using damage degree, with system is easy to use, flexible, cost is low, intelligence The advantages of degree is high can be changed, be suitable for equipping recovery force on a large scale.
(4) present invention carry out atural object damage degree assessment when used gray level co-occurrence matrixes image texture characteristic, Tamura textural characteristics, Gabor wavelet textural characteristics, straight line and circular image feature, close to the visual perception of human visual system, Damage degree of the earthquake to atural object can be preferably described, there is the advantages of calculating explicit physical meaning that is simple, calculating etc..
Brief description of the drawings
Fig. 1 is the method calculation flow chart of the present invention;
Fig. 2 is UAS operating diagram of the present invention.
Embodiment
As shown in figure 1, the assessment system that the inventive method is based on includes multi-rotor unmanned aerial vehicle subsystem, unmanned plane during flying Ground controls and output display subsystem composition, wherein:
Multi-rotor unmanned aerial vehicle subsystem is by multi-rotor unmanned aerial vehicle, two-freedom rotary motion camera, d GPS locating module, sky Middle wireless data transceiving terminal composition;Installed in multi-rotor unmanned aerial vehicle two-freedom rotary motion camera, d GPS locating module and On-air radio data transceiver terminal;Multi-rotor unmanned aerial vehicle be 4 rotors, 6 rotors or 8 rotors unmanned plane during flying device, more rotors without Man-machine to use battery powered, the non-stop flight time is not less than 30 minutes, and flying height is not less than 15 meters;Two-freedom rotary motion Camera can realize the rotation of -5 ° of pitching~185 °, 0 °~300 ° angular ranges of going off course;D GPS locating module is used to obtain nobody The current spatial coordinated information of machine;On-air radio data transceiver terminal is used to receive the control of unmanned plane during flying ground and output display The flight control instruction that subsystem is sent, and the image that the camera of multi-rotor unmanned aerial vehicle is shot sends back the unmanned plane on ground and flown Row ground controls and output display subsystem;
Unmanned plane during flying ground controls and output display subsystem is by terrestrial wireless data transceiver terminal and portable computing Machine forms;Terrestrial wireless data transceiver terminal be used for and multi-rotor unmanned aerial vehicle enter row data communication, control, monitor more rotors nobody The normal flight of machine;Portable computer is used for the picture of output display multi-rotor unmanned aerial vehicle shooting, and portable computer utilizes The grader of typical earthquake disaster image feature data set pair SVMs is trained, and according to the image of shooting using branch The grader for holding vector machine interacts the estimation of formula earthquake damage degree, the work shape of simultaneous real-time monitoring multi-rotor unmanned aerial vehicle State.
The present invention uses IEEE 802.11G wireless network transmissions agreement, and according to MPEG4 compressed format, will revolve more The image of wing unmanned plane camera shooting is controlled to unmanned plane during flying ground and output display subsystem is returned.Two-freedom revolves It is Visible Light Camera to turn moving camera.
As shown in Fig. 2 the earthquake disaster damage degree appraisal procedure based on unmanned plane of the present invention realizes step such as Under:
(1) typical earthquake disaster image feature data collection is established, the typical earthquake disaster image feature data collection includes House, road, bridge, dykes and dams, massif, river course and the earthquake disaster of vegetation image feature data collection:
To the multiple earthquake disaster images obtained based on unmanned plane in the past, determined by way of user's hand labeled Go out the image-region of the image-region acquisition more than 200 houses in house, the image-region in each house to determining calculates The characteristics of image in house, and damage degree in house corresponding to the image-region in each house is provided, so as to obtain the earthquake disaster in house Evil image feature data collection;
To the multiple earthquake disaster images obtained based on unmanned plane in the past, determined by way of user's hand labeled Go out image-region of the image-region acquisition more than 200 roads of road, the image-region of each road to determining calculates The characteristics of image of road, and road damage degree corresponding to the image-region of each road is provided, so as to obtain the earthquake disaster of road Evil image feature data collection;
To the multiple earthquake disaster images obtained based on unmanned plane in the past, determined by way of user's hand labeled Go out image-region of the image-region acquisition more than 200 bridges of bridge, the figure of bridge is calculated to the image-region of each bridge As feature, and bridge damage degree corresponding to the image-region of each bridge is provided, so as to obtain the earthquake disaster image of bridge spy Levy data set;
To the multiple earthquake disaster images obtained based on unmanned plane in the past, determined by way of user's hand labeled Go out image-region of the image-region acquisition more than 200 dykes and dams of dykes and dams, the figure of dykes and dams is calculated to the image-region of each dykes and dams As feature, and dykes and dams damage degree corresponding to the image-region of each dykes and dams is provided, so as to obtain the earthquake disaster image of dykes and dams spy Levy data set;
To the multiple earthquake disaster images obtained based on unmanned plane in the past, determined by way of user's hand labeled Go out image-region of the image-region acquisition more than 200 massifs of massif, the figure of massif is calculated to the image-region of each massif As feature, and massif damage degree corresponding to the image-region of each massif is provided, so as to obtain the earthquake disaster image of massif spy Levy data set;
To the multiple earthquake disaster images obtained based on unmanned plane in the past, determined by way of user's hand labeled Go out the image-region of the image-region acquisition more than 200 river courses in river course, the figure in river course is calculated to the image-region in each river course As feature, and damage degree in river course corresponding to the image-region in each river course is provided, so as to obtain the earthquake disaster image in river course spy Levy data set;
To the multiple earthquake disaster images obtained based on unmanned plane in the past, determined by way of user's hand labeled Go out image-region of the image-region acquisition more than 200 vegetation of vegetation, the figure of vegetation is calculated to the image-region of each vegetation As feature, and vegetation damage degree corresponding to the image-region of each vegetation is provided, so as to obtain the earthquake disaster image of vegetation spy Levy data set.
It is as follows for above-mentioned 7 quasi-representative image category, the mode of user's hand labeled:
For road, bridge, dykes and dams and river course, corresponding image-region is determined by way of user's hand labeled Method is as follows:The edge placement of road, bridge, dykes and dams and river course is marked by the way of straight line and Drawing of Curve, is then extracted Image-region is as the image-region determined between straight line and/or curve;For house, by way of user's hand labeled The method for determining the image-region in house is as follows:Draw a circle to approve out the position in house by hand using quadrangle, then extract quadrangle Internal image-region is as the image-region determined;For massif and vegetation, determined by way of user's hand labeled The method for going out corresponding image-region is as follows:Draw a circle to approve out the region of massif and vegetation by hand using polygon, then extract polygon Image-region inside shape is as the image-region determined.
Each characteristics of image of the present invention is defined as follows:
I) gray level co-occurrence matrixes angular second moment GLCMASM, the index characterization image texture intensity profile uniformity coefficient and line The metric of fineness is managed, its calculating formula is as shown in formula (1):
Wherein, p (i, j) is the gray level co-occurrence matrixes of piece image, and coordinate (i, j) represents image location information.
Ii) gray level co-occurrence matrixes the moment of inertia GLCMCON, the index characterization definition of image texture and the depth of texture Metric, its calculating formula is as shown in formula (2):
Iii) gray level co-occurrence matrixes correlation GLCMCOR, the index characterization measurement of texture similar area directionality size Value, its calculating formula is as shown in formula (3):
Wherein,Respectively WithAverage and mean square deviation, NgFor the maximum of single pixel gray level.
Iv) gray level co-occurrence matrixes entropy GLCMENT, the complexity or mixed and disorderly degree of index characterization gradation of image distribution, its Calculating formula is as shown in formula (4):
V) gray level co-occurrence matrixes inverse difference moment GLCMIDM, the uniformity coefficient of the index characterization image texture, its calculating formula such as formula (5) shown in:
Vi) the contrast T in Tamura textural characteristicscon, the intensity of the index characterization variation of image grayscale, it is counted Formula is as shown in formula (6) and (7):
Wherein μ4The fourth central for being image patch away from, σ is the mean square deviation for the image-region determined, α4For intermediate variable. When actually calculating, the rectangular window region of fixed size in the image-region determined, i.e. image patch are taken first, and according to From top to bottom, order from left to right determines region using the rectangular window area coverage image, and carries out in image quadravalence The heart is away from the calculating with mean square deviation.
Vii) Gabor direction characters fgabor:If g (x, y) represents Gabor functions, generating function is used as using g (x, y), is passed through Yardstick expansion and rotation transformation are carried out to g (x, y), one group of self similarity wave filter g can be obtainedmn(x, y), Gabor wavelet conversion can be determined Justice is formula (12), and corresponding textural characteristics can be calculated as formula (13) and (14).Gabor wavelet feature vector may be defined as formula (15)。
gmn(x, y)=a-mg(x′,y′) (9)
X '=a-m(xcosθ+ysinθ) (10)
Y '=a-m(-xsinθ+ycosθ) (11)
μmn=∫ ∫ | Wmn(x,y)|dxdy (13)
Wherein, g (x, y) represents Two-Dimensional Gabor Wavelets basic function, and W is Gaussian function multiple modulation frequency, σx、σyFor variable x With y mean square deviation;gmn(x, y) is self similarity wave filter, a-mIt is scale factor, a>1, x ', y ' are the calculating after being converted to x, y As a result, m and n is integer, represents corresponding yardstick and direction, and m ∈ [0, M-1], M are all dimensions, n ∈ [0, K-1], θ =n π K, K are all direction numbers, and subscript " * " represents conjugate complex number, x1、y1Expression Gabor wavelet integral transformation offset, I (x, Y) image-region selected out, W are representedmnThe result that (x, y) converts for Gabor wavelet, μmn、σmnFor Gabor wavelet average with Mean square deviation feature.
Viii) in the artificial structure of unmanned plane, be easier to occur the edge of building rule, as straight line, The basic elements such as circle, therefore the straight line number H of straight-line detection can be realized using Hough transform technologylWith the circle of loop truss Number HcFeature description calculate.With other lines compared with loop truss technology, the detection technique of Hough transform have amount of calculation it is small, Calculate the characteristics of accuracy is high.
Above-mentioned calculated disaster characteristic is formed into typical earthquake disaster image feature data collection, and it is special according to earthquake disaster The opinion of family, the grade destroyed according to disaster carry out the mark of disaster damage degree.A kind of typical earthquake disaster image feature data Collect sample as shown in 1 table.Damage degree concept proposed by the invention, mainly from the angle of graphical analysis come describe house, The destructiveness that road, bridge, massif etc. are subjected to when by earthquake or the influence of its secondary disaster.The index should be with being evaluated Ground artificial building or natural scene construction quality or geologic structure it is relevant, therefore with than indexs such as earthquake magnitude, degree of splitting more Specifically, accurately description effect more directly perceived.For example the place A and place B in somewhere simultaneously shake, what seismic detector measured Earthquake magnitude is similarly 6.0 grades;But the geologic structure on A ground is more firm, and culture's mass is preferable, when 6.0 grades of earthquakes occur, ground The damage that table is subjected to be not it is very serious, caused by casualties it is naturally smaller, the image result taken photo by plane is earth's surface culture It is little with natural scene damage degree;And B be located in mountain region along the river, earth's surface culture is second-rate, when 6.0 grades of ground occur There occurs serious earth's surface during shake to cave in, and culture all damages, and casualties is serious, the result taken photo by plane be culture with Natural scene damage degree is huge.The definition of damage degree be according to earthquake to caused by atural object destructiveness according to seismological expert and The artificial experience of rescue expert is defined.It is enumerated a kind of 5 grades of appraisal procedures of building construction damage degree in table 2, other 6 It is similar therewith that the damage degree of type atural object defines method.When the damage degree of reality is assessed, it is necessary to which seismological expert and rescue are special Family observes substantial amounts of historical earthquake and taken photo by plane data, and according to the content defined in table 2, atural object damage degree is carried out according to subjective judgement Differentiate and give a mark.Therefore, by the definition of image damage degree, the different Disaster degree in description two places that can be vivid and quantitative, Guidance is provided for follow-up rescue work.
The earthquake disaster damage degree of table 1 assesses image feature data collection sample
The house damage degree of table 2 defines method
Sequence number Damage situation describes Grade
1 Crackle occurs for surface structures, but does not cave in One-level
2 Severe crack occurs for surface structures, but does not cave in Two level
3 Surface structures occurs part and caved in, but major structural damage does not occur Three-level
4 Surface structures caves in, but structure is generation gross distortion Level Four
5 Surface structures caves in completely, and comminuted destruction occurs for building structure Pyatyi
When the Earthquake Secondary calamity that landslide secondary disaster, mud-rock flow secondary disaster, surface collapse, sedimentation and cracking occurs During evil, calculation and the training data and monitoring data that are obtained during above-mentioned generation earthquake disaster are completely the same;And due to all kinds of Disaster destructiveness to caused by ground is different, the span difference, therefore pass through above-mentioned side of corresponding each feature Method can realize the differentiation to all kinds of disasters.
(2) typical earthquake disaster image feature data collection training grader is used.
According to the data set accumulated in above-mentioned steps (1), the training of SVM classifier is carried out.SVM classifier is based on system The mathematical tool of the theories of learning is counted, compared with other grader instruments, SVM is more suitable for solving small sample, non-linear, higher-dimension Degree, the data of local minimum problem, the present invention determine that SVM classifier is more adapted to solve herein by attempting various graders Disaster damage degree classification problem.Need to train 7 graders, respectively house, road, bridge altogether in the implementation procedure of training Beam, dykes and dams, massif, river course, the image analysis result of vegetation damage situation, the indicator combination mode of graphical analysis can be found in table 1 In content.The training data of each grader is the various combination (referring to table 1) of image texture and edge feature, the prison of grader Superintend and direct the evaluation grade that data integrate damage degree for each disaster that seismological expert provides according to professional knowledge (referring to table 2).Each SVM The sample size of classifier training data should be more than 200 groups of data (positive and negative sample number is respectively more than 100), and final each SVM divides The classification precision of prediction of class device should be greater than 95%.Once classifier training is completed, then the output of the grader be then to earthquake and The description of its secondary disaster damage degree.
(3) new earthquake disaster image is obtained based on unmanned plane
Remote control multi-rotor unmanned aerial vehicle is flown or hovered in disaster area overhead, and during flight, carry is in nothing The camera of man-machine lower end carries out captured in real-time to the ground in disaster area, among the result of shooting is stored in into the internal memory of system, And as desired by the data taken photo by plane using IEEE 802.11G wireless network transmissions agreement and according to MPEG4 compressed format Among sending back the control of unmanned plane during flying ground and the output display subsystem on ground in real time, new earthquake disaster image is obtained.
(4) assessment of damage degree is carried out
To the new earthquake disaster image of acquisition, respectively for house, road, bridge by way of user's hand labeled Beam, dykes and dams, massif, river course, the quasi-representative atural object of vegetation 7 determine corresponding image-region, according to the image-region meter determined Characteristics of image corresponding to every quasi-representative atural object is calculated, the last characteristics of image according to corresponding to per quasi-representative atural object uses corresponding support The grader of vector machine carries out the assessment of damage degree.
The inventive method be more suitable for occur earthquake and secondary landslide, mud-rock flow, surface collapse, sedimentation with The disaster information that early stage is carried out during the disaster that ftractures is collected and assessed.
Unspecified part of the present invention belongs to general knowledge as well known to those skilled in the art.

Claims (8)

1. a kind of earthquake disaster damage degree appraisal procedure based on unmanned plane, it is characterised in that realize that step is as follows:
(1) establish typical earthquake disaster image feature data collection, the typical earthquake disaster image feature data collection include house, Road, bridge, dykes and dams, massif, river course and the earthquake disaster of vegetation image feature data collection;
To the multiple earthquake disaster images obtained based on unmanned plane in the past, room is determined by way of user's hand labeled The image-region in room obtains the image-region more than 200 houses, and the image-region in each house to determining calculates house Characteristics of image, and damage degree in house corresponding to the image-region in each house is provided, so as to obtain the earthquake disaster figure in house As characteristic data set;
To the multiple earthquake disaster images obtained based on unmanned plane in the past, determine to engage in this profession by way of user's hand labeled The image-region on road obtains the image-region more than 200 roads, and the image-region of each road to determining calculates road Characteristics of image, and road damage degree corresponding to the image-region of each road is provided, so as to obtain the earthquake disaster figure of road As characteristic data set;
To the multiple earthquake disaster images obtained based on unmanned plane in the past, bridge is determined by way of user's hand labeled The image-region of beam obtains the image-region more than 200 bridges, and the image spy of bridge is calculated to the image-region of each bridge Sign, and bridge damage degree corresponding to the image-region of each bridge is provided, so as to obtain the earthquake disaster characteristics of image number of bridge According to collection;
To the multiple earthquake disaster images obtained based on unmanned plane in the past, dike is determined by way of user's hand labeled The image-region on dam obtains the image-region more than 200 dykes and dams, and the image spy of dykes and dams is calculated to the image-region of each dykes and dams Sign, and dykes and dams damage degree corresponding to the image-region of each dykes and dams is provided, so as to obtain the earthquake disaster characteristics of image number of dykes and dams According to collection;
To the multiple earthquake disaster images obtained based on unmanned plane in the past, determine to come out of retirement and take up an official post by way of user's hand labeled The image-region of body obtains the image-region more than 200 massifs, and the image spy of massif is calculated to the image-region of each massif Sign, and massif damage degree corresponding to the image-region of each massif is provided, so as to obtain the earthquake disaster characteristics of image number of massif According to collection;
To the multiple earthquake disaster images obtained based on unmanned plane in the past, river is determined by way of user's hand labeled The image-region in road obtains the image-region more than 200 river courses, and the image spy in river course is calculated to the image-region in each river course Sign, and damage degree in river course corresponding to the image-region in each river course is provided, so as to obtain the earthquake disaster characteristics of image number in river course According to collection;
To the multiple earthquake disaster images obtained based on unmanned plane in the past, determine to plant by way of user's hand labeled The image-region of quilt obtains the image-region more than 200 vegetation, and the image spy of vegetation is calculated to the image-region of each vegetation Sign, and vegetation damage degree corresponding to the image-region of each vegetation is provided, so as to obtain the earthquake disaster characteristics of image number of vegetation According to collection;
The characteristics of image in house includes gray level co-occurrence matrixes angular second moment index GLCMASM, gray level co-occurrence matrixes index of correlation GLCMCOR, gray level co-occurrence matrixes entropy index GLCMENT, contrast index T in Tamura textural characteristicscon, Gabor direction characters Index fgabor, utilize Hough transform carry out straight-line detection straight line number index HlLoop truss is carried out with using Hough transform Circle number index Hc
Road, bridge, the characteristics of image in dykes and dams and river course include gray level co-occurrence matrixes angular second moment index GLCMASM, gray scale it is common Raw matrix inertia square index GLCMCON, gray level co-occurrence matrixes entropy index GLCMENT, gray level co-occurrence matrixes inverse difference moment index GLCMIDM、 Contrast index T in Tamura textural characteristicscon, Gabor direction character indexs fgaborStraight line is carried out with using Hough transform The straight line number index H of detectionl
Massif and the characteristics of image of vegetation include gray level co-occurrence matrixes angular second moment index GLCMASM, gray level co-occurrence matrixes inertia Square index GLCMCON, gray level co-occurrence matrixes inverse difference moment index GLCMIDM, contrast index T in Tamura textural characteristicsconWith Gabor direction character indexs fgabor
(2) it is each using house, road, bridge, dykes and dams, massif, river course and the training of the earthquake disaster of vegetation image feature data collection From support vector machine classifier;
(3) new earthquake disaster image is obtained based on unmanned plane;
(4) to the new earthquake disaster image of acquisition by way of user's hand labeled respectively for house, road, bridge, Dykes and dams, massif, river course, the quasi-representative atural object of vegetation 7 determine corresponding image-region, are calculated according to the image-region determined every Characteristics of image corresponding to quasi-representative atural object, supporting vector corresponding to the last characteristics of image use according to corresponding to per quasi-representative atural object The grader of machine carries out the assessment per quasi-representative atural object damage degree.
2. a kind of earthquake disaster damage degree appraisal procedure based on unmanned plane according to claim 1, its feature exist In:The earthquake disaster includes earthquake and its secondary disaster, and secondary disaster includes landslide, mud-rock flow, surface collapse, sedimentation With cracking.
3. a kind of earthquake disaster damage degree appraisal procedure based on unmanned plane according to claim 1, its feature exist In:For road, bridge, dykes and dams and river course, the method that corresponding image-region is determined by way of user's hand labeled It is as follows:The edge placement of road, bridge, dykes and dams and river course is marked by the way of straight line or Drawing of Curve, then extracts two Straight line, the image-region between two curves or straight line and a curve is as the image-region determined.
4. a kind of earthquake disaster damage degree appraisal procedure based on unmanned plane according to claim 1, its feature exist In:For house, the method that the image-region in house is determined by way of user's hand labeled is as follows:Using quadrangle hand Work draws a circle to approve out the position in house, then extracts the image-region inside quadrangle as the image-region determined.
5. a kind of earthquake disaster damage degree appraisal procedure based on unmanned plane according to claim 1, its feature exist In:For massif and vegetation, the method that corresponding image-region is determined by way of user's hand labeled is as follows:Using more Side shape draws a circle to approve out the region of massif and vegetation by hand, then extracts the image-region of polygonal internal as the image district determined Domain.
6. a kind of earthquake disaster damage degree appraisal procedure based on unmanned plane according to claim 1, its feature exist In:
<mrow> <msub> <mi>GLCM</mi> <mi>ASM</mi> </msub> <mo>=</mo> <munder> <mi>&amp;Sigma;</mi> <mi>i</mi> </munder> <munder> <mi>&amp;Sigma;</mi> <mi>j</mi> </munder> <mi>p</mi> <msup> <mrow> <mo>(</mo> <mi>i</mi> <mo>,</mo> <mi>j</mi> <mo>)</mo> </mrow> <mn>2</mn> </msup> </mrow>
<mrow> <msub> <mi>GLCM</mi> <mi>CON</mi> </msub> <mo>=</mo> <munder> <mi>&amp;Sigma;</mi> <mi>i</mi> </munder> <munder> <mi>&amp;Sigma;</mi> <mi>j</mi> </munder> <msup> <mrow> <mo>(</mo> <mi>i</mi> <mo>-</mo> <mi>j</mi> <mo>)</mo> </mrow> <mn>2</mn> </msup> <mi>p</mi> <mrow> <mo>(</mo> <mi>i</mi> <mo>,</mo> <mi>j</mi> <mo>)</mo> </mrow> </mrow>
<mrow> <msub> <mi>GLCM</mi> <mi>COR</mi> </msub> <mo>=</mo> <mfrac> <mrow> <munder> <mi>&amp;Sigma;</mi> <mi>i</mi> </munder> <munder> <mi>&amp;Sigma;</mi> <mi>j</mi> </munder> <mrow> <mo>(</mo> <mi>ij</mi> <mo>)</mo> </mrow> <mi>p</mi> <mrow> <mo>(</mo> <mi>i</mi> <mo>,</mo> <mi>j</mi> <mo>)</mo> </mrow> <mo>-</mo> <msubsup> <mi>&amp;mu;</mi> <mi>x</mi> <mi>GLCM</mi> </msubsup> <msubsup> <mi>&amp;mu;</mi> <mi>y</mi> <mi>GLCM</mi> </msubsup> </mrow> <mrow> <msubsup> <mi>&amp;sigma;</mi> <mi>x</mi> <mi>GLCM</mi> </msubsup> <msubsup> <mi>&amp;sigma;</mi> <mi>y</mi> <mi>GLCM</mi> </msubsup> </mrow> </mfrac> </mrow>
<mrow> <msub> <mi>GLCM</mi> <mi>ENT</mi> </msub> <mo>=</mo> <munder> <mrow> <mo>-</mo> <mi>&amp;Sigma;</mi> </mrow> <mi>i</mi> </munder> <munder> <mi>&amp;Sigma;</mi> <mi>j</mi> </munder> <mi>p</mi> <mrow> <mo>(</mo> <mi>i</mi> <mo>,</mo> <mi>j</mi> <mo>)</mo> </mrow> <mi>log</mi> <mo>[</mo> <mi>p</mi> <mrow> <mo>(</mo> <mi>i</mi> <mo>,</mo> <mi>j</mi> <mo>)</mo> </mrow> <mo>]</mo> </mrow>
<mrow> <msub> <mi>GLCM</mi> <mi>IDM</mi> </msub> <mo>=</mo> <munder> <mi>&amp;Sigma;</mi> <mi>i</mi> </munder> <munder> <mi>&amp;Sigma;</mi> <mi>j</mi> </munder> <mfrac> <mn>1</mn> <mrow> <mn>1</mn> <mo>+</mo> <msup> <mrow> <mo>(</mo> <mi>i</mi> <mo>-</mo> <mi>j</mi> <mo>)</mo> </mrow> <mn>2</mn> </msup> </mrow> </mfrac> <mi>p</mi> <mrow> <mo>(</mo> <mi>i</mi> <mo>,</mo> <mi>j</mi> <mo>)</mo> </mrow> </mrow>
Wherein, p (i, j) is the gray level co-occurrence matrixes for the image-region determined, coordinate (i, j) represents the position 2 for the image-region determined Confidence ceases, and is respectively <mrow> <msub> <mi>p</mi> <mi>x</mi> </msub> <mrow> <mo>(</mo> <mi>i</mi> <mo>)</mo> </mrow> <mo>=</mo> <munder> <mi>&amp;Sigma;</mi> <mi>k</mi> </munder> <mi>p</mi> <mrow> <mo>(</mo> <mi>i</mi> <mo>,</mo> <mi>k</mi> <mo>)</mo> </mrow> </mrow> Sum Average and mean square deviation, i=1,2 ..., maximum that Ng, j=1,2 ..., Ng, Ng are single pixel gray level.
7. a kind of earthquake disaster damage degree appraisal procedure based on unmanned plane according to claim 1, its feature exist In:
<mrow> <msub> <mi>T</mi> <mi>con</mi> </msub> <mo>=</mo> <mfrac> <mi>&amp;sigma;</mi> <msubsup> <mi>&amp;alpha;</mi> <mn>4</mn> <mrow> <mn>1</mn> <mo>/</mo> <mn>4</mn> </mrow> </msubsup> </mfrac> </mrow>
<mrow> <msub> <mi>&amp;alpha;</mi> <mn>4</mn> </msub> <mo>=</mo> <mfrac> <msub> <mi>&amp;mu;</mi> <mn>4</mn> </msub> <msup> <mi>&amp;sigma;</mi> <mn>4</mn> </msup> </mfrac> </mrow>
Wherein μ4It is the fourth central square of image patch, σ is the mean square deviation for the image-region determined, α4For intermediate variable.
8. a kind of earthquake disaster damage degree appraisal procedure based on unmanned plane according to claim 1, its feature exist In:Gabor direction character indexs fgaborCalculation formula it is as follows:
<mrow> <mi>g</mi> <mrow> <mo>(</mo> <mi>x</mi> <mo>,</mo> <mi>y</mi> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mn>1</mn> <mrow> <mn>2</mn> <mi>&amp;pi;</mi> <msub> <mi>&amp;sigma;</mi> <mi>x</mi> </msub> <msub> <mi>&amp;sigma;</mi> <mi>y</mi> </msub> </mrow> </mfrac> <mi>exp</mi> <mo>[</mo> <mo>-</mo> <mfrac> <mn>1</mn> <mn>2</mn> </mfrac> <mrow> <mo>(</mo> <mfrac> <msup> <mi>x</mi> <mn>2</mn> </msup> <msubsup> <mi>&amp;sigma;</mi> <mi>x</mi> <mn>2</mn> </msubsup> </mfrac> <mo>+</mo> <mfrac> <msup> <mi>y</mi> <mn>2</mn> </msup> <msubsup> <mi>&amp;sigma;</mi> <mi>y</mi> <mn>2</mn> </msubsup> </mfrac> <mo>)</mo> </mrow> <mo>+</mo> <mn>2</mn> <mi>&amp;pi;jWx</mi> <mo>]</mo> </mrow>
gmn(x, y)=a-mg(x′,y′)
X '=a-m(xcosθ+ysinθ)
Y '=a-m(-xsinθ+ycosθ)
<mrow> <msub> <mi>W</mi> <mi>mn</mi> </msub> <mrow> <mo>(</mo> <mi>x</mi> <mo>,</mo> <mi>y</mi> <mo>)</mo> </mrow> <mo>=</mo> <mo>&amp;Integral;</mo> <mo>&amp;Integral;</mo> <mi>I</mi> <mrow> <mo>(</mo> <mi>x</mi> <mo>,</mo> <mi>y</mi> <mo>)</mo> </mrow> <msubsup> <mi>g</mi> <mi>mn</mi> <mo>*</mo> </msubsup> <mrow> <mo>(</mo> <mi>x</mi> <mo>-</mo> <msub> <mi>x</mi> <mn>1</mn> </msub> <mo>,</mo> <mi>y</mi> <mo>-</mo> <msub> <mi>y</mi> <mn>1</mn> </msub> <mo>)</mo> </mrow> <mi>d</mi> <msub> <mi>x</mi> <mn>1</mn> </msub> <mi>d</mi> <msub> <mi>y</mi> <mn>1</mn> </msub> </mrow>
μmn=∫ ∫ | Wmn(x,y)|dxdy
<mrow> <msub> <mi>&amp;sigma;</mi> <mi>mn</mi> </msub> <mo>=</mo> <msqrt> <mo>&amp;Integral;</mo> <mo>&amp;Integral;</mo> <msup> <mrow> <mo>[</mo> <mo>|</mo> <msub> <mi>W</mi> <mi>mn</mi> </msub> <mrow> <mo>(</mo> <mi>x</mi> <mo>,</mo> <mi>y</mi> <mo>)</mo> </mrow> <mo>|</mo> <mo>-</mo> <msub> <mi>&amp;mu;</mi> <mi>mn</mi> </msub> <mo>]</mo> </mrow> <mn>2</mn> </msup> <mi>dxdy</mi> </msqrt> </mrow>
Wherein, g (x, y) represents Two-Dimensional Gabor Wavelets basic function, and W is Gaussian function multiple modulation frequency, σx、σyFor variable x and y Mean square deviation;gmn(x, y) is self similarity wave filter, a-mIt is scale factor, a>1, x ', y ' are to the result of calculation after x, y conversion, m It is integer with n, represents corresponding yardstick and direction, m ∈ [0, M-1], M are all dimensions, n ∈ [0, K-1], θ=n π/K, K is all direction numbers, and subscript " * " represents conjugate complex number, x1、y1Gabor wavelet integral transformation offset is represented, I (x, y) is represented The image intensity of (x, y) opening position, WmnThe result that (x, y) converts for Gabor wavelet, μmn、σmnFor Gabor wavelet average with Mean square deviation feature.
CN201410454210.8A 2014-09-05 2014-09-05 A kind of earthquake disaster damage degree appraisal procedure based on unmanned plane Active CN104239885B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201410454210.8A CN104239885B (en) 2014-09-05 2014-09-05 A kind of earthquake disaster damage degree appraisal procedure based on unmanned plane

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201410454210.8A CN104239885B (en) 2014-09-05 2014-09-05 A kind of earthquake disaster damage degree appraisal procedure based on unmanned plane

Publications (2)

Publication Number Publication Date
CN104239885A CN104239885A (en) 2014-12-24
CN104239885B true CN104239885B (en) 2017-11-28

Family

ID=52227909

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201410454210.8A Active CN104239885B (en) 2014-09-05 2014-09-05 A kind of earthquake disaster damage degree appraisal procedure based on unmanned plane

Country Status (1)

Country Link
CN (1) CN104239885B (en)

Families Citing this family (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104915495B (en) * 2015-06-05 2017-09-05 中国科学院水利部成都山地灾害与环境研究所 A kind of Debris-flow Hazards appraisal procedure and application
CN105096315B (en) * 2015-06-19 2018-03-06 西安电子科技大学 Heterogeneous super-pixel SAR image segmentation method based on Gamma distributions
WO2017173563A1 (en) * 2016-04-05 2017-10-12 盛玉伟 Method and system for detecting real estate quality
JP6458782B2 (en) * 2016-07-28 2019-01-30 カシオ計算機株式会社 Display control apparatus, display control method, and program
CN106295699A (en) * 2016-08-11 2017-01-04 北京师范大学 A kind of earthquake Damage assessment method and apparatus based on high-definition remote sensing data
CN108154067A (en) * 2016-12-02 2018-06-12 航天星图科技(北京)有限公司 A kind of mud-rock flow area monitoring method
CN106969751B (en) * 2017-03-13 2017-11-14 西安科技大学 A kind of method of the coal mining subsidence amount monitoring calculation based on unmanned aerial vehicle remote sensing
CN108304809B (en) * 2018-02-06 2020-03-27 清华大学 Near real-time earthquake damage assessment method based on post-earthquake aerial image
CN109145812A (en) * 2018-08-20 2019-01-04 贵州宜行智通科技有限公司 Squatter building monitoring method and device
CN109544579A (en) * 2018-11-01 2019-03-29 上海理工大学 A method of damage building is assessed after carrying out calamity using unmanned plane

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1828336A (en) * 2006-04-04 2006-09-06 张路平 Rapid estimation method for earthquake disaster damage based on GIS technology
CN101979961A (en) * 2010-05-18 2011-02-23 中国地震局地球物理研究所 Disaster condition acquisition system
CN103034863A (en) * 2012-12-24 2013-04-10 重庆市勘测院 Remote-sensing image road acquisition method combined with kernel Fisher and multi-scale extraction
CN103034870A (en) * 2012-12-14 2013-04-10 南京思创信息技术有限公司 Ship fast identification method based on features
CN103198333A (en) * 2013-04-15 2013-07-10 中国科学院电子学研究所 Automatic semantic labeling method of high resolution remote sensing image
CN103426158A (en) * 2012-05-17 2013-12-04 中国科学院电子学研究所 Method for detecting two-time-phase remote sensing image change
US8625890B1 (en) * 2011-10-17 2014-01-07 Google Inc. Stylizing geographic features in photographic images based on image content
CN103699903A (en) * 2013-12-24 2014-04-02 中国科学院深圳先进技术研究院 City roof green area calculation method and system based on image identification
CN103761740A (en) * 2014-01-23 2014-04-30 武汉大学 Construction damage assessment method based on single post-earthquake POLSAR image

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1828336A (en) * 2006-04-04 2006-09-06 张路平 Rapid estimation method for earthquake disaster damage based on GIS technology
CN101979961A (en) * 2010-05-18 2011-02-23 中国地震局地球物理研究所 Disaster condition acquisition system
US8625890B1 (en) * 2011-10-17 2014-01-07 Google Inc. Stylizing geographic features in photographic images based on image content
CN103426158A (en) * 2012-05-17 2013-12-04 中国科学院电子学研究所 Method for detecting two-time-phase remote sensing image change
CN103034870A (en) * 2012-12-14 2013-04-10 南京思创信息技术有限公司 Ship fast identification method based on features
CN103034863A (en) * 2012-12-24 2013-04-10 重庆市勘测院 Remote-sensing image road acquisition method combined with kernel Fisher and multi-scale extraction
CN103198333A (en) * 2013-04-15 2013-07-10 中国科学院电子学研究所 Automatic semantic labeling method of high resolution remote sensing image
CN103699903A (en) * 2013-12-24 2014-04-02 中国科学院深圳先进技术研究院 City roof green area calculation method and system based on image identification
CN103761740A (en) * 2014-01-23 2014-04-30 武汉大学 Construction damage assessment method based on single post-earthquake POLSAR image

Also Published As

Publication number Publication date
CN104239885A (en) 2014-12-24

Similar Documents

Publication Publication Date Title
Cook An evaluation of the effectiveness of low-cost UAVs and structure from motion for geomorphic change detection
Yang et al. Spatial and spectral hybrid image classification for rice lodging assessment through UAV imagery
Lisein et al. A photogrammetric workflow for the creation of a forest canopy height model from small unmanned aerial system imagery
Mui et al. An object-based approach to delineate wetlands across landscapes of varied disturbance with high spatial resolution satellite imagery
Vollgger et al. Mapping folds and fractures in basement and cover rocks using UAV photogrammetry, Cape Liptrap and Cape Paterson, Victoria, Australia
Nurminen et al. Performance of dense digital surface models based on image matching in the estimation of plot-level forest variables
Lucieer et al. Mapping landslide displacements using Structure from Motion (SfM) and image correlation of multi-temporal UAV photography
Lucieer et al. Using an Unmanned Aerial Vehicle (UAV) to capture micro-topography of Antarctic moss beds
CN104616093B (en) A kind of seismic disaster relief command dispatching system and method
Jakubowski et al. Tradeoffs between lidar pulse density and forest measurement accuracy
CN104931022B (en) Satellite image stereoblock adjustment method based on spaceborne laser altimeter system data
Al-Rawabdeh et al. Using an unmanned aerial vehicle-based digital imaging system to derive a 3D point cloud for landslide scarp recognition
Jaboyedoff et al. New insight techniques to analyze rock-slope relief using DEM and 3Dimaging cloud points: COLTOP-3D software
Lu et al. Land cover classification in a complex urban-rural landscape with QuickBird imagery
Woodget et al. Subaerial gravel size measurement using topographic data derived from a UAV‐SfM approach
CN106127725B (en) A kind of millimetre-wave radar cloud atlas dividing method based on multiresolution CNN
Shahbazi et al. Recent applications of unmanned aerial imagery in natural resource management
Laliberte et al. Image processing and classification procedures for analysis of sub-decimeter imagery acquired with an unmanned aircraft over arid rangelands
Pirotti et al. Suitability of LiDAR point density and derived landform curvature maps for channel network extraction
Liao et al. High-resolution remote sensing mapping of global land water
Ajayi et al. Generation of accurate digital elevation models from UAV acquired low percentage overlapping images
Elberink et al. The use of anisotropic height texture measures for the segmentation of airborne laser scanner data
Kavzoglu et al. Parameter-based performance analysis of object-based image analysis using aerial and Quikbird-2 images
CN100390565C (en) Method for measuring forest by unmanned aerial vehicle aerial photography remote sensing
Razak et al. Airborne laser scanning of forested landslides characterization: Terrain model quality and visualization

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant